Datasets:

jbrinkma

/

pile-2m

like 0

Mitchells Presbyterian Church Mitchells Presbyterian Church is a historic Presbyterian church located on VA 652 in Mitchells, Culpeper County, Virginia. It was built in 1879, and is a one-story, frame building in the Carpenter Gothic style. It measures 50 feet by 30 feet and sits on a brick foundation. The interior features a trompe-l'œil fresco added between 1892 and 1899 by well-known local artist Joseph Oddenino. He also painted the interior murals at Elmwood. It was listed on the National Register of Historic Places in 1980. References Category:Churches on the National Register of Historic Places in Virginia Category:Presbyterian churches in Virginia Category:National Register of Historic Places in Culpeper County, Virginia Category:Carpenter Gothic church buildings in Virginia Category:Churches completed in 1879 Category:Churches in Culpeper County, Virginia

Q: People says machine understands binary code, that is 0 and 1 is that literally true? What i meant to say that 0 and 1 is still a perceptive number that human to understand so how come machine understands this? If machine can get 0 and 1 it can equally understand alphabets also. So where i am missing the concept? A: Machines don't actually understand a 1 or a 0 as a number the way we do. What we mean by 1 or 0 (in the context of machine language) is really just a high or a low voltage value stored in some sort of charge storage structure (often a MOS capacitor these days). A: On a very low level, microcontrollers/microprocessors are made of numerous transistors acting as switches. You can think of each transistor as only having two states: "on" and "off". These transistors are driven by voltages. If we apply a voltage, we get a specific behavior. If we remove the voltage, we get the other behavior. We need a convenient way to say whether a transistor has a voltage applied or not. It is customary to say "1" if there is an applied voltage, or "0" if there is none. So, computers don't actually understand "zeros and ones", as we think of them. They operate based on where the voltages are applied. But since we refer to these voltage states as "zeros and ones", the language usage makes sense.

Starting last week, Skype users began to complain on the company’s forum that they were being locked out from accessing the popular voice over IP service. As complaints mounted, it was guessed that Skype was no longer operational on Macs running older versions of OS X. Microsoft today confirmed the change in service in an answer to a question regarding Skype compatibility with OS X 10.6 Snow Leopard, which will for now remain unaffected. Skype message board staff member Rene writes: Hi everyone, there seems to be a lot of confusion around this. First, let me assure you that no-one is locking you out from using Skype on OS X 10.6 – 10.8 and there is no need to upgrade to Mavericks or (the new OS X now in dev preview) Yosemite if you don’t want to. All OS X 10.6 – 10.8 users can head over to http://www.skype.com/en/download-skype/skype-for-mac/downloading/ and get version 6.15. This is a very solid release and you can still use all of the main features without any problems (i.e. group video, group chats, screen sharing etc). Note that you should only use the link above, it serves you with a correct update for your OS version, direct links (to DMG files) can work, but are not guaranteed to do so. That said, we still encourage you to stay up to date, meaning upgrading to Mavericks and getting the latest Skype. Hope this clarifies things a bit. If you have problems updating, please let us know in this thread and I’ll try to help you. Current versions of Skype require a Mac featuring an Intel-based processor. For those with older Macs, industrious users have fashioned workarounds, including a method of blocking the software’s update mechanism from connecting with update servers, though these bootstrapped solutions are hit-or-miss. In June, Microsoft said it would “retire” older versions of Skype for both Mac and Windows over the next few months, forcing current users to upgrade to the latest builds for uninterrupted service. At the time, the company noted users with Skype for Mac (6.13 and below) and Skype for Windows (6.14 and below) would be unable to log in until they update. The latest version of Skype for Mac is 6.19 and which can be downloaded for free from the company’s website.

CHICAGO - Sophomore forward Lauren Bohaboy (Mission Viejo, Calif./Santa Margarita) scored with 4:52 left in the second overtime to lift No. RV/24 Notre Dame to a thrilling 1-0 BIG EAST Conference victory at DePaul on a rainy and cold Friday afternoon at Wish Field in Chicago. The Fighting Irish win, coupled with Georgetown's loss by the same score at Villanova earlier in the day, gave Notre Dame a share of the BIG EAST National Division title, marking the 14th time the Fighting Irish have earned a regular-season division or overall championship since joining the BIG EAST in 1995. Notre Dame also is now unbeaten in 10 of its last 11 matches. Freshman forward Crystal Thomas (Elgin, Ill./Wheaton Academy) and sophomore defender Taylor Schneider (Southlake, Texas/Carroll Senior) both were credited with assists on Bohaboy's match-winning goal, which was the first overtime winner for Notre Dame since Nov. 6, 2009 (Jessica Schuveiller vs. St. John's in BIG EAST Championship semifinals at Storrs, Conn.), and the latest OT goal for the Fighting Irish since Sept. 21, 2001 (Amy Warner at 114:11 against Villanova at old Alumni Field, back when the two overtime periods were 15 minutes in length, instead of the current 10-minute frames). Notre Dame (12-4-2, 8-1-1 BIG EAST) dominated the statistical breakdown of Friday's match at DePaul, holding a 20-5 edge in total shots, with three of the Blue Demons' shots coming in a two-minute flurry at the end of the first half. The Fighting Irish also piled up a 7-3 margin in shots on goal and a 5-1 advantage on corner kicks, while fouls were 13-8 against Notre Dame (as well as the lone yellow card that went to Thomas late in the first overtime). Freshman goalkeeper Elyse Hight (Edmond, Okla./Bishop McGuinness) wasn't called upon for much action on Friday, but she did end up with three saves, including a huge point-blank stop on DePaul's Rachel Pitman in the 44th minute. Meanwhile, Megan Pyrz registered six saves in the Blue Demon goal, including a pair of leaping stops that helped keep the hosts in contention until the waning moments of added time. "For the most part, I thought it was a pretty complete game for us," Notre Dame head coach Randy Waldrum said. "We maintained possession well, which is really hard to do on a tight slick field like this, but we need to do better in the final third. A lot of the credit for that has to go to their defense, which was strong, but Lauren (Bohaboy) came up with a nice finish to give us what I thought was a just result. "It's a great reward for this young team to earn a piece of the BIG EAST division championship," he added. "These players continue to work hard in training every day and they're starting to see the positive results that come from that development. We're not where we need to be yet, but this was one of our goals before the season and we were able to achieve it, which always makes a coach proud." Chances were at a premium throughout Friday's match, due in part to DePaul's defensive-minded approach. Still, Notre Dame had several opportunities to break through, beginning in the 19th minute, when Bohaboy won a free kick 20 yards out, just off the top left corner of the box. However, her curling drive over the wall towards the upper left corner was pushed over the bar by Pyrz. The Fighting Irish had another enticing chance in the 39th minute, as freshman midfielder Glory Williams (Dallas, Texas/Lake Highlands) found classmate and defender Stephanie Campo (Basking Ridge, N.J./Ridge) with a well-weighted chip down the left flank. Campo ran on to the ball and one-timed a service towards the far right post, where junior midfielder/tri-captain Elizabeth Tucker (Jacksonville, Fla./Bishop Kenny) drove a sharp header on frame, only to be denied by a retreating Pyrz. DePaul's best looks of the afternoon came in the closing minutes of the first half, starting from a corner kick by Natasa Radosavljevic, whose service from the right flag caromed off a defender and landed near the penalty spot. Pitman alertly tried to snap a quick shot off, but Hight was equal to the task, going down to make the reflex stop, and Kelly Blumberg's 10-yard shot off the rebound sailed harmlessly over the bar (43:45). The second half and first overtime was more of a back-and-forth affair, with each side making runs into the attacking third, but neither able to create any meaningful scoring chances. Williams had perhaps the most dangerous look of the second half with nine minutes left, testing Pyrz with a 30-yard drive from the center channel but the ball drifted a couple of feet over the top. After a largely unrewarding first overtime, Notre Dame found the solution midway through the second extra period. Schneider collected a ball on the right flank 30 yards out and calmly drove a service into the box, to the right of the penalty spot. Thomas settled the cross, battling past a pair of DePaul defenders and muscling the ball over to Bohaboy. The second-year striker then made no mistake, hammering a low shot from 12 yards out inside the left post past a diving Pyrz, sending the Fighting Irish reserves racing on to the pitch in celebration. As the No. 2 seed in the BIG EAST's National Division (by virtue of its loss at Georgetown on Oct. 12), Notre Dame will kick off postseason play at 1 p.m. (ET) Oct. 28, when it plays host to Syracuse (No. 3 seed in American Division) in a BIG EAST Championship quarterfinal match at Alumni Stadium. POST-MATCH NOTES: Notre Dame earns its first overtime win since Nov. 6, 2009, when it defeated St. John's, 2-1 in a BIG EAST Championship semifinal at Storrs, Conn., on Jessica Schuveiller's goal (and Rose Augustin's assist) at 98:42 -- coincidentally, Augustin was in the stands at Wish Field on Friday, as part of a large contingent of enthusiastic Notre Dame alumni and supporters ... Bohaboy's goal was the latest overtime score for the Fighting Irish since Sept. 21, 2001, when Amy Warner scored at 114:11 to defeat Villanova, 2-1 at old Alumni Field (in what was the program's first match in 12 days following the Sept. 11 terrorist attacks on the United States) ... the Fighting Irish are 21-6-16 (.674) in overtime matches during the 14-year Waldrum era (1999-present) ... Notre Dame improves to 8-0-0 all-time against DePaul (4-0-0 at Wish Field) with a 28-1 all-time series scoring margin, although the past three visits to Chicago have seen the Fighting Irish win low-scoring affairs (1-0 in 2008 and 2012, 2-0 in 2010) ... Schneider's assist was her first point of her two-year college career, and made her the 16th different Notre Dame player to register a point this season.

Background ========== Discarding gallbladder specimens without histopathological analysis is best avoided. Selective approach for sending these specimens to the laboratory results in missing discrete pathologies like premalignant benign lesions such as porcelain gallbladder, carcinoma-in-situ and early carcinomas \[[@B1],[@B2]\]. Early carcinoma of gallbladder notoriously remains undiagnosed without histopathology as it neither produces clinical symptoms or signs nor provides any clues on ultrasound assessment. Cholecystectomy performed with provisional diagnosis of benign diseases based on clinical, ultrasonological and computerized tomographic scanning misses a significant number of early malignant lesions of gallbladder. To avoid such blunders with bad consequences, therefore, every cholecystectomy specimen should be routinely examined histologically \[[@B3]\]. Carcinoma gallbladder carries one of the worst of cancer mortality. It is a frequent underlying pathology in patients undergoing surgical intervention for chronic cholecystitis with cholelithiasis \[[@B4]\]; long standing chronic inflammation by gallstones is considered an important etiological role in carcinogenesis. The incidence of carcinoma gallbladder associated with gallstones varies from 0.3 to 12 per cent \[[@B5],[@B6]\]. Histopathological analysis is therefore mandatory for diagnosis of early carcinomas. Good prognosis is anticipated in patients with gallbladder carcinoma discovered as an incidental finding in stage I disease \[[@B5]\]. The practice of discarding gallbladder specimen is standard in most tertiary care hospitals of Pakistan including the primary investigator's own institution on the pretext that "surgeon knows best which gallbladder is to be sent to laboratory". Histopathology is restricted to only those specimens, which show gross abnormalities. This practice is based on the assumption that gallbladder carcinoma is always associated with macroscopic abnormalities. At the same time, this selective approach is justified by claiming that it reduces patient's financial liabilities and pathologist's workload. This contradicts to the worldwide practice where gallbladder specimen is invariably sent for histological analysis for the sole purpose of identifying discrete carcinoma in early stage. This study was conducted to assess the feasibility or otherwise of performing histopathology in every specimen of gallbladder. This would ensure picking of discrete carcinoma of gallbladder, hidden to both the human eyes and touch, which in turn will assist in decreasing mortality rate. Methods ======= This cohort study was carried out at the Department of Surgery, Liaquat University Hospital, Hyderabad/Jamshoro, Pakistan, over a period of two years from March 2010 to February 2012. The study was designed in compliance with the Helsinki Declaration and a formal approval for the study was obtained from the institutional Ethics Review Committee (ERC), Liaquat University of Medical & Health Sciences. The study consisted of 220 patients with acute or chronic cholecystitis secondary to gallstones admitted through the outpatient Department. Non-probability convenient sampling technique was used. Written, informed consent was taken from all the participants and/or their next of kin where so applied. Patients with evidence of carcinoma gallbladder, on clinical grounds and confirmed on ultrasonography and/or CT scan were excluded. Gallbladders showing gross abnormalities suggestive of localized or infiltrative malignancy during surgery were excluded. Detailed history and thorough clinical examination of the patients was done with special attention to the right hypochondrium for preoperative assessment of palpable mass. Systemic review was done to see any co-morbidity. Baseline and specific investigations especially ultrasonography of abdomen was done in all patients. Laparoscopic cholecystectomy was performed in all cases but had to be converted to open procedure in few cases where anatomical distortion and dense adhesions prevented any further progress laparoscopically. All gallbladder specimens, including those with no obvious gross abnormality, were sent for histopathology. A predesigned proforma was used to put down the information gathered. Data were analyzed using SPSS version 17.0. Results ======= Over a period of two years, two hundred and twenty patients with symptomatic gallstones were admitted for cholecystectomy. There were 193 females and 27 males with a male to female ratio of 1:7. The age ranged from 19 to 80 years with the mean age of 32.3 ± 5.3 years (Table [1](#T1){ref-type="table"}). Majority of patients (92%) presented with upper abdominal pain of varying duration. Other symptoms are depicted in Table [2](#T2){ref-type="table"}. ###### **Age distribution**\* **Age of patients(years)** **No.of Patients(n=220)** **%** ---------------------------- --------------------------- ------- **19**-**30** 54 24.6 **31**-**40** 61 27.8 **41**-**50** 59 26.8 **51**-**60** 34 15.4 **61**-**70** 10 4.5 **71**-**80** 02 0.9 \*Mean age 32.25±5.3 years. ###### Presenting symptoms **Symptoms** **No.of Patients(n=220)** **%** ----------------------------- --------------------------- ------- Pain in Upper abdomen 201 91.4 Intolerance to fatty food 134 61 Nausea and/or vomiting 44 20 Mass in right hypochondrium 5 2.3 All 220 gallbladders were palpated and were opened per-operatively for any focal or diffuse thickening of the gallbladder wall, a raised mucosal plaque, polypoid growth or an infiltrating grey white mass. The specimens were then sent for histopathology. Two hundred and three of the specimens showed evidence chronic cholecystitis, 7 acute cholecystitis with mucocele, 3 acute cholecystitis with empyema and one incidental associated polyp. Six gallbladders (2.8%) showed evidence of adenocarcinoma of varying differentiation along with cholelithiasis, (Table [3](#T3){ref-type="table"}). ###### Histopathological report (n=220) **Histopathology** **Male** **Female** **%** ----------------------------------- ---------- ------------ --------- Chronic Cholecystitis 18 185 92.3 Acute cholecystitis with mucocele 5 2 3.2 Acute cholecystitis with empyema 2 1 1.3 Polyp 0 1 0.5 Malignancy 2 4 2.7 **Total** **27** **203** **100** There were six incidental carcinomas with no gross abnormalities in this series. Subsequent staging revealed three adenocarcinomas in stage T1b, two in stages T2 and one in T3 stage. Three patients with T1 tumors did not undergo any further procedure; three patients with advanced stages (T2 and T3) underwent revision surgery with resection of liver segments 4b and 5 (T2) and right hepatectomy (T3) along with lymphadenectomy. Port site rim of about 2 cms were also excised and sent for histopathology and was found to be free of tumor seedling in all three cases. Discussion ========== In this study, females outnumbered males with male to female ratio of 1:7. Female predominance is also reported by similar studies \[[@B7],[@B8]\]. The mean age 32.25±5.3 years ranging from 19 to 80 years, slightly higher than that reported in other studies \[[@B9]\]. Over ninety one per cent patients presented with pain upper abdomen, a number significantly lower than that reported by Laghari et al \[[@B10]\]. where all patients had upper abdominal pain. None of the patients in our study had any evidence of malignancy either clinically or on ultrasound examination. The most common histopathological finding in our study was chronic cholecystitis; 203 (92.3%) specimens were reported as chronic inflammation with mucosal ulceration, denudation, metaplasia to dysplasia and wall infiltration by chronic inflammatory cells like neutrophils, macrophages, plasma cells and varying degrees of fibrosis. A similar study by Memon \[[@B11]\] also reports chronic cholecystitis as major histopathological finding, identified in 64.8% cases. Empyema of the gallbladder is often difficult to distinguish from uncomplicated acute cholecystitis \[[@B12]\]. In this study, 3 (1.3%) cases were reported as acute cholecystitis with empyema of gallbladder. This is in stark contrast to 31.5% cases of empyema associated with cholecystitis as reported by Memon \[[@B11]\]. Mucocele of the gallbladder has an incidence of 3 percent. In this study, 3.2 cases presented as cholecystitis with mucocele; reported incidence of mucocele is, however, several times higher \[[@B13],[@B14]\]. Gallbladder polyps have an incidence ranging from 4.6 to 6.9 per cent \[[@B15]\]. In our study, one case of gallbladder polyp was identified. This low incidence can be attributed to small number of cases in our series. The only polyp discovered in our study was in a female. However, the prevalence of this pathology is much higher amongst males \[[@B16]\]. In our series, incidental carcinoma of gallbladder was found in 6 cases (2.7%). These gallbladders showed no gross abnormality per-operatively. The incidence of gallbladder malignancy in this series was considerably low compared to other studies, which show an incidence varying from 6.9 to 12 per cent \[[@B6],[@B17],[@B18]\]. The low incidence of malignancy in our series can be attributed to high sensitivity to exclusion criteria, where all patients with any preoperative evidence of malignancy, no matter how trivial, were excluded from the study. Low incidence of malignancy in our patients can also be attributed to increased acceptance and early reporting for laparoscopic cholecystectomy due to resort to day-care surgery at our institution. Samad \[[@B3]\] reports an incidence of 1.1% of malignancy in patients who underwent cholecystectomy for presumed chronic cholecystitis with cholelithiasis. All patients in our series presented with longstanding history of chronic cholecystitis. There were no symptoms or signs suggestive of underlying malignancy in any patient; gallbladder malignancy usually does not have any characteristic clinical features with over 90 per cent of patients presenting with symptoms of acute or chronic cholecystitis \[[@B19]\]. Although ultrasound has a high diagnostic accuracy for both advanced and early gallbladder cancer \[[@B20]\], none of the six carcinomas in this series were picked on preoperative ultrasound. In addition, all six gallbladder specimens showed no macroscopic evidence of malignancy when they were opened during surgery. This is in contrast to the study by De Zoyasa et al. in whom all four cancers were suspected either on preoperative ultrasound or grossly during surgery; they suggest a more selective approach to gallbladder histology which may have saved both time and cost without having any unfavorable effects on patients well-being \[[@B21]\]. Similar observations and recommendations are made by other studies \[[@B22],[@B23]\]. The issue of routine histopathology of all gallbladder specimen therefore remains unresolved; the need to send every specimen for histopathology or otherwise therefore depends on the expertise of the ultrasonologist as it depends on the skill of the operating surgeon. We, however, advocate routine histopathology of all gallbladders removed at surgery since the subsequent report would provide evidence of malignancy on solid grounds. Although there are myriad of premalignant conditions, carcinoma gallbladder has a strong association with gallstones \[[@B17]\]. The strong association between the two warrants attention paid to histopathology of specimen in all cases undergoing cholecystectomy for cholelithiasis, irrespective of presence or otherwise of any gross abnormalities. It is widely reported that long standing mucosal irritation by the stones cause atypical cellular changes and increased cellular proliferation. It has been hypothesized that in long standing cases, these areas of hyperplasia progress to metaplasia and carcinoma-in- situ \[[@B24]\]. Studies confirm presence of such changes in the vicinity of gallbladder carcinoma \[[@B25]\]. The incidence of malignancy in our study was 2.7%, which is similar to that reported by Khan AH \[[@B26]\]. Histopathology revealed adenocarcinoma of gallbladder in six patients. One tumor was in stage T1b, two in T2 while one was in stage T3. While Cholecystectomy is an adequate treatment for pT1 tumors, pT2 and pT3 tumors require revision surgery to achieve a tumor-free surgical margin, along with lymph node dissection \[[@B27]\]. Three patients with the unexpected gallbladder cancer in stages T2 and T3 underwent a second standard revision procedure including transection of liver segments 4b and 5 (T2) and right hepatectomy (T3) with lymphadenectomy. These patients were followed up the oncological department for a period varying from 6 months to one year before being lost to follow up. All six cases of incidentally diagnosed malignancies had associated gallstones, thereby strongly supporting the role of chronic irritation by long standing gallstones as etiological factor for carcinoma gall bladder. Grading of cancer was found well differentiated to poorly differentiated adenocarcinoma, no case of squamous cell carcinoma, or other variant of cancerous histopathology seen. Conclusion ========== The histopathological spectrum of gallbladder after cholecystectomy is extremely variable. Incidental diagnosis of carcinoma gallbladder is not rare; we discovered evidence of malignancy in 6 (2.8%) cases on subsequent histopathological examination of gallbladder specimen, which showed no gross features of cancer. These cases had no symptoms suggestive of underlying malignancy nor was cancer reported on any of the preoperative investigations. Had the protocol of routine histopathology of all gallbladder specimens not been followed, these cases would have been failed to identify with disastrous results; We therefore, strongly advocate routine histopathology of all cholecystectomy specimens. Old patients and patients having long duration symptoms are strong candidates for histopathology of gallbladder specimen. The limitations of this study included a small sample size. Competing interests =================== The authors declare that they have no competing interests. Authors' contributions ====================== FGS carried out acquisition, analysis, interpretation of the data and drafting of the manuscript. AAM was involved interpretation of the data, drafting of the manuscript, and revised it critically for the intellectual content till the final version was reached. AHA, NAS and LA have read, edited and approved the final manuscript. All authors read and approved the final manuscript. Pre-publication history ======================= The pre-publication history for this paper can be accessed here: <http://www.biomedcentral.com/1471-2482/13/26/prepub> Acknowledgement =============== The authors acknowledge the contribution of Prof. Abdul Sattar Memon, Dean of Surgery & Allied Sciences, Liaquat University of Medical & Health Sciences Jamshoro for critical and final revision of the manuscript before its submission. The study was self-funded.

Q: Why did the Ottomans rebuild Byzantine gates to a smaller format? It's very perplexing to me. Here are some examples: https://pages.pomona.edu/~sg064747/travel/images/Turkey/Fortifications/1600-LandWallsArchGate.jpg https://romeonrome.com/files/2013/09/66.EmperorsGoldenGate-600x450.jpg https://pages.pomona.edu/~sg064747/travel/images/Turkey/Fortifications/1024-YedikuleGoldenGate.jpg http://romeartlover.tripod.com/2a.jpg http://www.livius.org/site/assets/files/21526/thumbnail_istanbul_wall_theodosius_02_ab.jpg https://media-cdn.tripadvisor.com/media/photo-s/0a/0f/da/40/walls-of-constantinople.jpg One theory I heard about is that it's due to cultural differences between the conquerors and the conquered. A: Out of the list of images in the question, some can be found on this site, which provides some descriptions of the photos. One of these is identified as the 'Golden Gate' of (then) Constantinople. The Wikipedia entry says this about the Golden Gate: It was the main ceremonial entrance into the capital, used especially for the occasions of a triumphal entry of an emperor into the capital on the occasion of military victories or other state occasions such as coronations. These military victories became much fewer, and finally: With the progressive decline in Byzantium's military fortunes, the gates were walled up and reduced in size in the later Palaiologan period, and the complex converted into a citadel and refuge. The Palaiologan period was the last period of Byzantine rule, so the premise of the question is off. The gates were actually walled in by the Byzantines as a defensive measure at a time when Constantinople was more often the target of attack. The Ottomans did not close this gate in.

From 5 Stars to First Rounders, the Predictably Uncertain Path from Recruit to Pro Before signing day I took a look at how team recruiting rankings were predictive of future success. I found that good defenses almost always come with good recruits, but on offense great offense often comes without being fully stocked, although it doesn’t hurt. This week I wanted to look more at the individual level by comparing recruiting rankings to draft success. For most positions college success is going to translate well into future draft status. Michigan might have the biggest exception to that rule in Denard Robinson (although some think he might be a top WR pick). For almost everywhere on the field but rushing quarterback, college success and production are highly correlated to NFL stock. It’s not perfect but it’s a great place to start. The debate on do recruit rankings matter rages on. Dr. Saturday, may he blog in peace, annually refreshed his look to affirm their accuracy. Rarely do you find anything resembling an analytical take down but from even the best writers on college football can come the anecdotal dismissal. Hopefully those of us who prefer to use data have already won you over and this can be a nice look at some of the ups and downs within the overall success of recruiting rankings. If you’re there yet, hopefully you are after you read this. The Data Sets On the recruit side, the pool of players will be the recruiting classes of 2002-2006. All but 2-3 of those players have had their shot to be drafted between the 2005 and the 2011 drafts. I will only be looking at the players who were ranked for their position, as well. This means I have all 4 & 5 stars and the best of the 3 stars. I excluded fullbacks and specialists because the numbers are pretty low and they are mostly all 3 stars or less. It’s All in How You Word It There are two key arguments against recruiting rankings. The first is the one used by Bruce Feldman in his recent article on Stanford linked above. It’s the yeah but what about…argument. Ignore recruiting rankings because Stanford is good. Ignore recruiting rankings because JJ Watt is good. There of course exceptions. There are plenty of flameouts and come from nowhere success stories but this is a volume game and the exceptions don’t disprove the rule. The second argument is the famed failure to divide. Here are two true statements: If you are drafted, you are more likely to be a three star or less recruit than four or five star. The more stars you have the more likely you are to be drafted. The first statement is used by opponents of rankings but isn’t really a relevant statement. The second is the key point. If every single five star was drafted, there would still be six times more three stars and below drafted than five stars. Because four stars and above are so selective they can’t win the quantity game but they dominate the likelihood game. The NFL is full of unheralded recruits but for every five start there are literally hundreds of unheralded recruits playing college football. The pool just starts much bigger. Tell Me Something I Don’t Know So at this point we can all agree that recruiting rankings matter, right? If you’ve made it this far you’ve earned a chart. Percent of Recruits Drafted Position* 5 star 4 star 3 star APB 50% 10% 9% ATH 62% 22% 9% C NA 11% 6% CB 56% 31% 11% DT 64% 21% 12% G 50% 15% 10% ILB 64% 18% 8% OLB 22% 23% 13% PQB 44% 21% 9% RB 43% 21% 8% RQB 29% 14% 6% S 67% 15% 10% SDE 40% 22% 10% T 20% 20% 11% TE 100% 30% 13% WDE 30% 30% 11% WR 53% 22% 10% All 48% 21% 10% *Position based on recruited position, not drafted position Across all positions, each additional star more than doubles your likelihood of being drafted. It’s not only true in the aggregate but at the position level, as well. There isn’t a single position where a 3 star recruit is more likely to be drafted than a four star. And this is a self-selected group of 3 stars and not the entire pool. In almost every case, a fifth star is another large bump from 4 stars. OLB, OT and WDE are virtually equivalent between 4 and 5 stars. Even a largely college specific position like Dual-Threat QB (RQB) and undefined positions like Athlete show the same trend. The top positions for 5 star success are Athlete, DT, ILB and Safety at over 60% and the tight end position which was a perfect 4/4 in getting 5 stars drafted. But getting drafted is only half the story, the other is draft position. Average Pick For Drafted Players Position 5 star 4 star 3 star APB 9 66 158 ATH 111 102 144 C 108 84 CB 104 124 97 DT 120 102 119 G 82 124 137 ILB 33 124 118 OLB 125 118 118 PQB 39 98 88 RB 41 111 141 RQB 14 105 147 S 135 118 123 SDE 96 139 123 T 94 88 141 TE 63 114 124 WDE 53 70 124 WR 77 95 118 All 79 107 124 At the position level, the draft spot doesn’t hold up quite as well as the previous chart, but overall there is a strong trend favoring the higher starred players. On average, a drafted five star player will be picked in the middle of the third round, nearly a full round ahead of the average four star player and another 17 picks ahead of ranked three star players. On twitter on Friday I teased a question about which position did five stars underperform four star counterparts. There is actually a position on each side of the ball. On defense it’s outside linebackers that don’t follow the trend and on offense it’s the tackles. I think it’s interesting that Rivals has struggled to match top high school talent at position like tackle, outside linebacker and defensive end at the rate they have at other positions. Despite the weakness at these positions, similar positions like guard, inside linebacker and defensive tackle have had their rankings hold up quite well. The Takeaway Don’t get too hung up on the magic of the fourth or fifth star. They are a nice aggregation but there isn’t going to be much difference between the last five start and the first four star. The bottom line is the higher ranked a recruit is the better they are likely to be, with plenty of exceptions. Positions like tackle, weakside d-end and outside linebacker the difference between a four star and a five is almost negligible. And there are no guarantees. Loading up on top talent gives you the highest likelihood of having team success and successful individuals, but when you get down to the specific player level it becomes a crapshoot. More 5 stars players never hear their names called than ones who do. For four stars it’s still a nearly 4:1 chance against getting drafted. Or is Hoke using his gold pooping magic to defy those odds as well? I know Mattison joked that Hoke chose to coach the NT spot because Martin played that position, but my emotional thinking likes to believe it was to get BWC on the correct track after three years of, let's say, questionable coaching. The path from high school ranking to NFL draft position goes through a college program. The analysis I would like to see is on which programs are best at reducing high star flameout, and/or highest percentage of producing over-achievers. I would hypothesize that places like Alabama, where there is a disproportionate wealth of talent, also have a disproportionate amount of flameout or under performance from high school star to NFL draft position. All of this can't be black and white, (although media writing articles unfortunately play upon the readers who want their answers diluted to black and white) but I am interested in knowing which "lever" has a larger effect; star rating or program development. In terms of recruiting rankings predicting success--I suspect it has to do with the lack of guards being rated that highly. (Similar to how all five-star TEs have gotten drafted over the sample.) Because some positions are perpetually rated, when recruiting services do decide to rank an exemplary guard highly, it's because they have a high degree of certainty that said guard is freaking awesome. None of the 5 star player counts are that high, but guards were 3/6 when it came to 5 stars getting drafted and tackles were 2/10. It's not just a more tackles are rated question. But there is definitely uncertainty tied to the small sample size. I didn't bring it out as an issue in the article because even with the small sample size the results were very consistent. I'm a Rivals guy also, so I appreciated you using those numbers, but the time honored debate is if one of the services is better than the other. It may be interesting to compare specifically the 3 categories that came out negligible - tackle, weakside d-end and outside linebacker and see if Scout is indeed better when it comes to evaluating the talent. Thanks, Mathlete. Very interesting. One of my favorite things to do is go way back to the old recruiting rankings and see where NFL stars were ranked as high schoolers. Granted, the systems probably weren't as accurate and well-organized back then, but still very interesting. A few of the best players at their positions in the NFL (from 2002 on): Calvin Johnson: high 4 star, rivals 100 player. Aaron Rodgers: generic 3 star Adrian Peterson: #1 overall player in nation. Derrelle Revis: generic 3 star Mario Williams: low-end 4 star These are just a few examples of how widely it varies. I don't know football compared to most people on this site, I just find it fun to go back and take a look. In my mind, the 2 players who are hands down the best at their positions in the league are Megatron and Revis, and neither of them were 5 stars, although Calvin was ranked pretty high. Those are all interesting examples- but the whole point (as I see it) of compiling data like this is to see general trends. Many fans (and perhaps talent evaluators) get stuck up on examples that break from the norm to deny trends or to skew their perceptions of players- this data shows very clearly that the more starts you have the more likely you are to be drafted- with a couple unique counter-examples where for some positions 4* and 5* are not much different. Yes 3* will outperform and get drafted above 4* and 5*- but there's a reason we get so excited about the massive number of 4* recruits we are pulling in- they do outperform 3* players as is demonstrated by a whole wealth of data just shared with us. You're using Rivals ratings right? Have you tried matching up their numeric ratings to likelihood of being drafted? I'm curious if that would help give a more granular result? I also think it's also interesting to compare this to All-American rankings. While five star players are just under five times more likely to get drafted than three star players per your analysis, Dr. Saturday found that five stars were more than eleven times more likely to make an All-American team than three stars. That's a pretty significant gap. I can speculate about a couple reasons for why, but the most obvious is a preference for placing a 5 star onto such a team over a 3 star (while the draft is more of an obvious meritocracy). Awesome analysis. But... because recruiting rankings are not static (ie. They change after a recruit commits to a school) there is ample opportunity for selection bias. The only analysis that would separate such bias would be an evaluation of recru iting rankings' performance within each university's pool of players. Otherwise you are simply "proving" that players from big-time programs (with big time facilities, big time coaching, and big time media exposur e) are drafted at a higher rate than players from the lower level programs. We all know that a recruit at Michigan is more likely to go pro than a recruit at EMU. I would rather know if Michigan got better recruits than ohio. I was thinking much the same thing but more on the recruit side of things and not the evaluater side. It seems (with zero data to back this up) that 5 star recruits almost exclusively end up at big time programs that are at least regular league championship contenders. This leads to more national exposure which leads to more NFL scouts being aware of the players which leads to a higher likelyhood of being drafted. In the modern NFL, scouts are aware of good players no matter where they play. If a player's college career is overlooked by the NFL, it's more likely to be because the player had injury problems or wasn't a schematic fit in college. I think the big-time college advantage is more self-reinforcing than anything else. Better raw material goes to school with better resources--and, boom, you get a multiplicative effect. The advantage in facilities, nutrition, and coaching is probably pretty big between big-time schools and, say, MAC School X. Also, JimLahey looks at how some of the best professional players were rated coming out of high school. I thought I'd look at some of the best college players, so the Heisman trophy winners of the last 7 years (looking at Rivals rankings, which I can only go back to Bush): Looking at the breakdown, 3 5 stars, 3 4 stars, and 1 3 star in the group. Obviously, the scope of this is limited and skewed, because the best teams typically get the best talent around the higher rated recruits, but was interesting. A breakdown of All Americans and their ratings may be a bit more appropriate. i bet the correlation gets better every year. scouts are better able to evaluate players than they used to be because tape is easier to send and because word spreads faster about good players b/c of social media and the like. I had actually attempted to a very similar analysis about 18mo ago, but I gave up because I had too many things on my plate. What I was attempting to do was see which recruiting service (ESPN, Rivals, Scout) was the best at predicting success. In a nutshell, I was going to do a series of regressions to see which service was the best predictor. For my own curiousity (and hopefully others'), do you think you could so something like that at some point? I am working on it. Only Scout and Rivals have data far enough back to do a good analysis with the NFL draft because of the lag time. Getting it all uploaded and managed is a bear. For any of you NOTY fans, you can understand how hard it is to link up all the creative spellings and name changes (like Maruce (Jones-)Drew or Patrick Johnson (Peterson). At some point I am hoping to get somewhere with it but don't count on it any time soon. This would be very interesting but the only problem is that there are a variety of confounding variables that really can't be accounted for. For one, coaching staffs and facilities differ and it's possible that the same 3* who was drafted in Setup A would not have been in Setup B. Also, where a recruit goes to school may lead to heightened exposure that could improve their draft stock. I would be curious to see if the 3*s that are drafted in the first couple rounds come disproportionately from certain schools. just certain advantages/biases that a recruit who goes to a big program may have over a recruit at a smaller program (frankly, I think the NFL generally does too good a job scouting to let talented guys slip through the cracks just because they are at a small school). The bigger and more interesting question is whether the lower ranked players who go to big programs actually then to be better recruits than their similarly ranked counterparts. If Nick Saban is hot after a 3-star player, I suspect that kid may well be more talented than his ranking suggests, and perhaps Rivals just missed the boat. In other words, are all 3-stars created equal, or are some more likely to succeed than other? Would be really interesting to compare, for example, whether a RIvals 5.7 3-star who is an otherwise highly ranked class is more likely to be drafted than a Rivals 5.7 3-star who is in a middling class, based on the theory that a lower ranked guy in a highly ranked class is much more likely to be a guy the coaches really wanted, rather than a guy the coaches merely had to settle for. On the other hand, a lower ranked guy in a highly ranked class might be LESS likely to succeed because the competition for playing time is greater. Either way, it would be interesting to know if there is any correlation between the overall ranking of a recruiting class and the likelihood of a lower ranked player in that class getting drafted. Mathlete, is this a case where median draft position might be a better indicator than average draft position? I can see an argument either way, but I'd think the median would be less sensitive to distortions from effects like picks being driven in part by NFL teams' positions of greatest need, which would probably alter the average draft position for the top few players more than the average position for the rest. I don't think Rivals does a very good job of projecting OL positions or DEs. They rate almost everyone as an offensive tackle (even though you need just as many guards on the team, and lots of the "tackles" end up at guard), and they often leave guys ranked as defensive ends who are clearly too big to be edge players. I think kids see the big money that OTs and DEs make, and they hate the thought of moving inside to lesser renowned positions. They suggest the odds for a RANDOMLY selected 5, 4 or 3 star player. However, it doesn't, for example, tell you that a particular 4 star player that your coaches just signed has a better likelihood of success than the 3 star player they just signed. It doesn't mean they necessarily make a mistake (even putting needs aside) when they go hard after a lower ranked guy and ignore a higher ranked guy. Because at that point you are no longer dealing with a random sampling from the pool--you are dealing with players that were specifically selected because the coaches saw something they wanted. You cannot forget that end of the day the rankings are just some guy's opinion, and opinions--even qualified opinions--vary. One person's three star maybe another person's five star. On the other hand, while you expect some variation in opinions, you also expect a good deal of overlap. If your coaches are consistently pulling in ALL lower ranked players, it's POSSIBLE they are seeing something everyone else is missing, but it is far more likely they just aren't doing a very good job getting or identifying the best players available.

Q: 親子関係で再帰処理したい クラス内のメソッドで、下記関係が成立しているとき、再帰処理するには？ ・childには、「中身がある配列」か「空の配列」が格納されているのですが、「空の配列」になるまで処理継続したい public function a() //中略 foreach ($data as $k => $v) { $this->r[] = $k; $this->f($v); } } public function f($v) { $child = $v->child; foreach ($child as $k2 => $v2) { $this->r[] = $k2; $this->f2($v2); } } public function f2($v2) { $child = $v2->child; foreach ($child as $k3 => $v3) { $this->r[] = $k3; $this->f3($v3); } } public function f3($v3) //以下略 A: 再帰呼び出しとは、プログラムのある関数の中から自分自身の関数を呼び出すことである。 ご質問中のソースにいくつかのメソッドがあるんですが、実際にロジックは一緒ですので、下記のロジックに変更して見てください。 public void mainMethod(element) { if(element->field is Empty) { return; } for(ele in element->field) { mainMethod(ele); } } よろしくお願いします。

Ukraine scores own goal as Euro debacle looms 歐洲盃足球賽災難降臨 烏克蘭自踢烏龍球 A pub promotes the Euro 2012 soccer tournament in London, Britain on May 29.這張五月二十九日攝於倫敦的照片，是一間推銷二○一二歐洲盃足球賽的酒吧。 Photo: EPA照片：歐新社 Ukraine failed to anticipate the angry Western reaction to its treatment of jailed ex-prime minister Yulia Tymoshenko and the dispute risks ruining its hosting of the Euro 2012 soccer championship, observers said. The Euro 2012 tournament co-hosted with Poland was set to be a glorious showcase for the ex-Soviet country, but President Viktor Yanukovych appears to have scored a massive own goal by allowing it to be overshadowed by the Tymoshenko case. Austria has announced it will boycott all matches hosted by Ukraine, a move that reportedly could be matched by Germany. All EU commissioners will also be absent, while at least seven EU heads of state are shunning a summit to be hosted by Yanukovych in Yalta this month. Tymoshenko, who was jailed for seven years on charges of abuse of power in October, upped the stakes in her standoff with the authorities last month by going on hunger strike and claiming she had been beaten by prison guards. With an efficient PR machine fronted by the opposition leader’s telegenic, London School of Economics-educated daughter Yevgeniya, the Western reaction led by Germany has been tough, and it clearly caught the Yanukovych government off guard. “No one expected that Germany would kick up such a fuss,”said a Ukrainian source close to the presidency, admitting that the administration had not drawn up any response to the crisis.

A convenient strategy for the synthesis of 4,5-bis(o-haloaryl)isoxazoles A series of new 1,2-bis(o-haloaryl)ethanones is efficiently prepared and applied to the synthesis of 4,5-bis(o-haloaryl)isoxazoles. Isolation of intermediate hydroxyisoxazolines, which are structurally examined, provides a definitive proof for a heterocyclization mechanism based on an amine exchange process. The isolation and X-ray crystallographic studies of significant side products such as benzamides and triarylpropionitriles are also described.

About Aussies Abroad - Federici extends Reading deal Qantas Socceroos goalkeeper Adam Federici has signed a new deal with Reading, which will keep him with the English Premier League club until 2015. Qantas Socceroos goalkeeper Adam Federici has signed a new deal with Reading, which will keep him with the English Premier League club until 2015. Federici, whose contract was due to expire next year, said he was pleased to commit to the club, who won promotion to the Premier League at the end of last season. "I've been here a long time," said Federici, 27, who first joined the Royals in 2005. "I've put in a lot of blood, sweat and tears so it wouldn't have been right to do anything other than sign again." Reading manager Brian McDemott said: "He is a lesson to young professionals about what is required to get to the top of your game. His workrate, diligence and desire are second to none." Football Family members have the exclusive opportunity to purchase tickets to the Qantas Socceroos- next FIFA World Cup Qualifiers, both to be held in Sydney, against Oman on March 26 and Iraq on June 18. Presale are available now, until 9pm on Sunday, 5 August, with Football Family members to receive a 15 per cent discount for tickets to both matches, which will cost just $68, alternatively single match tickets start from just $36 at a 10 per cent discount. If you are not yet a Football Family member you can join for free by clicking here. Be the first to get your tickets and get behind the Qantas Socceroos as they head toward Brazil 2014! Click here for full ticket details.

//===----------------------------------------------------------------------===// // // The LLVM Compiler Infrastructure // // This file is dual licensed under the MIT and the University of Illinois Open // Source Licenses. See LICENSE.TXT for details. // //===----------------------------------------------------------------------===// // <utility> // template <class T1, class T2> struct pair // template<class U, class V> pair(U&& x, V&& y); #include <utility> #include <memory> #include <cassert> int main() { #ifndef _LIBCPP_HAS_NO_RVALUE_REFERENCES { typedef std::pair<std::unique_ptr<int>, short*> P; P p(std::unique_ptr<int>(new int(3)), nullptr); assert(*p.first == 3); assert(p.second == nullptr); } #endif // _LIBCPP_HAS_NO_RVALUE_REFERENCES }

Q: How to stop Javascript from repeating the function? Guys I have a javascript which has a each loop in which I am calling a particular function for each index value. This script executes on mouse scroll event upon reaching page end (lazy loading). This is the main section of the script : function (data) { $.each(data, function (index, value) { BindNotice(value); }); } Now the problem is, after the index reaches it max value, it restarts from the 0 index. So I am getting repeated data in my view. I want to restrict this script to only execute until the index has reached the max value. How would I go about doing this ? EDIT : ADDED COMPLETE FUNCTION function callMoreData() { $.ajax( { type: "GET", url: "/api/values/getnotice", dataType: "json", crossDomain: true, async: true, cache: false, allow: true, success: function (data) { $.each(data, function (index, value) { BindNotice(value); }); }, error: function (x, e) { alert('problem while fetching records!'); } } );} From JS file : $(".mainwrap .innnerwrap").mCustomScrollbar({ autoDraggerLength:true, autoHideScrollbar:true, scrollInertia:100, advanced:{ updateOnBrowserResize: true, updateOnContentResize: true, autoScrollOnFocus: false }, callbacks:{ whileScrolling:function(){WhileScrolling();}, onTotalScroll: function () { callMoreData(); } } }); A: You could call another function, and then set it to do nothing after first go through. function (data) { updateData(data); } function updateData(data) { updateData = function(data) {}; $.each(data, function (index, value) { BindNotice(value); }); } Demo using alerts. I made some edits to the code because it sounds more like you're getting multiple events and they are writing over previous events. After the first call, the function is basically set to do nothing. Thus the first callback will be the only callback that calls BindNotice. Edit As I mentioned in the comments, to ever receive data again through this channel, you need to restore updateData. Here's an example: function restoreUpdateData() { updateData = function(data) { updateData = function(data) {}; $.each(data, function (index, value) { BindNotice(value); }); }; }

Narcissus Narcissus is a local florist here in Edinburgh working with prestigious clients such as Harvey Nichols and many luxury jewellery brands. The branding needed to be understated enough to sit alongside the vast array of colours used in their floral arrangements but also had to feel luxury in order to appeal to their premium client base. The stem device was created as a subtle nod towards the world of botanics.

Filed 8/4/17 CERTIFIED FOR PUBLICATION IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA FIRST APPELLATE DISTRICT DIVISION FOUR ALFONSO AYALA, Plaintiff and Appellant, A142830 v. (Solano County RANDY DAWSON, Super. Ct. No. FCS040044) Defendant and Respondent. After living for more than a dozen years in a residential unit he claimed he owned, Alfonso Ayala was evicted by the property owner, Randy Dawson, in an unlawful detainer action. Ayala defended by attempting to quash service of summons on the ground he was not a tenant, but instead held equitable title under an oral installment sale contract for the purchase of property. Dawson countered that, in fact, Ayala was a tenant under a written lease, and had breached the lease in various ways, thus justifying his eviction. After an evidentiary hearing on Ayala’s motion to quash service, Dawson prevailed and ultimately took a default judgment. Ayala then vacated the premises. In this case, a separate, concurrent action by Ayala against Dawson for fraud and various other claims, Ayala once again pursues the theory that he holds equitable title under an installment sale contract. He seeks to argue, as he did in the unlawful detainer action, that Dawson, a real estate broker, deceived him into signing the lease, while misrepresenting that the document was simply the memorialization of a preexisting oral contract of sale. The court granted summary judgment for Dawson, ruling that, under the doctrine of collateral estoppel, Ayala is barred from relitigating his fraud-in-the- inducement theory. This appeal is from the ensuing judgment and from an award of 1 attorney fees in Dawson’s favor under the prevailing party fee clause in the lease. We affirm. I. BACKGROUND In late 1999, Ayala found a five-unit residential property in Vacaville that he wished to buy, but he could not qualify for a mortgage loan, so he sought the assistance of his friend Dawson, a real estate broker. According to Ayala, the two men struck an oral agreement under which Dawson agreed to obtain the required purchase money mortgage loan and buy the property in Dawson’s name for a price of $330,000; Ayala agreed to pay the entire down payment of 20 percent and thereafter pay Dawson a $200 per month fee, plus the monthly principal and interest on the mortgage loan; Ayala agreed to maintain the property and not to commit waste; Ayala agreed to contract with a property management company to manage the rental units; and, upon Ayala’s payoff of the entire principal and interest due on the mortgage, Dawson agreed to deed the property to Ayala in fee. The parties executed a written contract provided by Dawson on December 9, 1999, which Ayala—in reliance on Dawson’s superior knowledge of real estate transactions— claims he understood merely to confirm an installment contract for the purchase of property on terms the two men had previously discussed. After signing and initialing the papers, Ayala moved into one of the units and contends he spent hundreds of thousands of dollars improving the property over the more than 12 years he lived there. Ayala used a “shop” on the property in his business and may have stored equipment in a “barn” on the property. From January 2000 to 2008, he also paid Dawson $2,700 per month, and from 2008 to July 2012, he paid Dawson $2,900 per month. Payments to Dawson apparently were made by the property management company, which used rental income from the property, supplemented by funds provided by Ayala. Unbeknownst to Ayala— who testified he did not read the “pile of documents” presented to him in December 1999 —the contract he executed was a “RESIDENTIAL LEASE WITH OPTION TO PURCHASE (CALIFORNIA SHORT FORM),” not an installment sale contract. According to Dawson, Ayala allowed the purchase option granted under this lease to 2 expire on December 31, 2004, and thereafter Ayala’s interest in the property amounted to nothing more than a month-to-month tenancy. As evidence that both parties understood their relationship that way, Dawson points out he raised what he explicitly denominated as the monthly “rent” from $2,700 per month to $2,900 per month in 2008, without protest of any kind by Ayala. Despite the fact that the option to purchase had expired, in 2011 Dawson offered to sell the property to Ayala for the original price of $330,000, with a credit for the down payment he had paid. Ayala refused the offer, taking the position he should not have to buy property he already owned. By then, according to Dawson, Ayala had fallen into a pattern of late payment of his monthly rent and had allowed the property to fall into a state of disrepair. As a result, Dawson claimed, he was unable to obtain refinancing, which put him in a financial bind because he was in a “negative cash flow” position.1 According to Dawson, the whole idea for the transaction had been to provide, in effect, temporary financing so that Ayala could purchase the property no later than December 31, 2004, but that never happened, and after a decade, the continued carrying costs were damaging Dawson’s credit. Dawson testified that when he tried to raise these concerns with Ayala in August 2011, Ayala for the first time began trying to claim that he was not a tenant, that he held equitable title to the property under an oral installment sale contract, and that the written lease was fraudulent. In June 2012, Ayala filed this action, a lawsuit alleging claims against Dawson for fraud, breach of an oral contract, specific performance, preliminary and permanent injunctive relief, and declaratory relief. He later amended the complaint to add claims seeking a quiet title declaration, restitution for unjust enrichment, and the imposition of a constructive trust and a resulting trust. In July 2012, Dawson filed an action of his own 1 Apparently, the debt service on Dawson’s original loan had gone up because, in 2003, he initially refinanced the property, converting from a 30-year term loan to a 15- year term loan. His unsuccessful effort to refinance in 2011 appears to have been an effort to lower the debt service cost on the 15-year loan. 3 for unlawful detainer.2 Rather than answer, Ayala filed a motion to quash service of summons on the ground that the court in which the unlawful detainer action was filed lacked jurisdiction because “[t]here was never a landlord-tenant relationship between the parties.” In support of that jurisdictional argument, Ayala contended the written lease was fraudulent and Ayala occupied the property not as a tenant, but as a vendee-in- possession pursuant to an oral installment sale contract. After holding a one and a half day evidentiary hearing in the unlawful detainer action, Judge D. Scott Daniels denied the motion to quash, ruling that “[b]ased on Plaintiff’s . . . Residential Lease with Option to Purchase and the conduct of the parties, [Dawson had] met his burden of proof to establish the existence of a landlord-tenant relationship . . . .” In denying the motion, Judge Daniels found that Ayala had read and signed the lease, declined to find any basis to relieve Ayala of his contractual obligations, and specifically rejected the theory that Dawson served as Ayala’s real estate broker or was Ayala’s trusted real estate advisor. Ayala sought a writ of mandate in the superior court appellate division, requesting review of Judge Daniels’s ruling. The appellate division panel denied the writ, specifically affirming Judge Daniels’s finding of a landlord-tenant relationship as supported by substantial evidence. The panel also found no merit to a contention from Ayala that he had been deprived of due process because of the expedited handling of the hearing on his motion to quash. In addressing the due process issue, the writ panel stated “[t]he sole issue in the unlawful detainer action is possession and the arguments that [Ayala] wished to raise concerning ‘fraud, constructive trust, estoppel, breach of fiduciary duties and many integrally related factual matters’ will be exhaustively decided in the separate unlimited civil action [he has] filed,” referring to this action. 2 On May 21, 2015, Ayala filed a motion requesting that we take judicial notice of various pleadings filed in the unlawful detainer action under Evidence Code sections 451, 452 and 459 and California Rules of Court, rule 8.252. We grant the motion. 4  After affirmance of the denial of the motion to quash in Dawson’s unlawful detainer action and entry of judgment in that action by default,3 Ayala continued to pursue this action against Dawson. Discovery remained open in the case for nearly two years. In February 2014, Dawson filed a motion for summary judgment, arguing, among other things, that Judge Daniels’s rejection of Ayala’s fraud theory in the unlawful detainer action collaterally estops Ayala from pursuing the same theory in this action. Judge Daniels’s ruling, once it became final after Ayala exhausted his efforts to seek review, Dawson argued, is dispositive on every one of the claims asserted in this case. In opposition, Ayala argued once again that he had not been given enough time to prepare for the hearing on the motion to quash or for meaningful discovery, but his opposition papers identified no witnesses who had not testified before Judge Daniels in the unlawful detainer action and presented no documents that had not been considered by Judge Daniels. Judge Scott Kays granted Dawson’s motion for summary judgment in May 2014. He found that “[Dawson] has shown that [Ayala] is unable to establish an essential element of any of [his] causes of action because [he] is precluded from relitigating the validity and enforceability of the parties’ written lease agreement”; that “[a]ll of [Ayala’s] causes of action are based on [his] allegations that [Dawson] improperly memorialized the parties’ oral agreement for an installment sales contract as a lease agreement with an option to purchase . . . [which Dawson] fraudulently induced [Ayala] to execute . . . by misrepresenting the contents of the agreement . . .”; and that “the same parties have already fully litigated the validity and enforceability of the lease agreement in a prior unlawful detainer action.” Judgment pursuant to the order granting summary judgment was entered June 11, 2014. An attorney fees award for $27,000 in favor of Dawson under a 3 Ayala filed a notice of appeal from the judgment against him in the unlawful detainer action, but ultimately abandoned that appeal. 5 prevailing party fee clause in the lease issued later, on July 31, 2014. Ayala timely appealed both from the judgment and the attorney fees award. II. DISCUSSION A. Standard of Review and General Principles of Res Judicata The central issue here is whether the claims Ayala seeks to bring in this action are barred under principles of res judicata by the denial of his motion to quash in Dawson’s unlawful detainer action. Except for evidence as to which objections have been made and sustained, we take as true the facts shown by the evidence presented in opposition to a summary judgment motion, reviewing the grant of summary judgment de novo. (In re Automobile Antitrust Cases I and II (2016) 1 Cal.App.5th 127, 150–151.) “The defense of res judicata not only is properly raised by a motion for summary judgment, but also is a proper ground upon which to grant a summary judgment.” (Rohrbasser v. Lederer (1986) 179 Cal.App.3d 290, 296 (Rohrbasser).) “Whether the doctrine of res judicata applies in a particular case is a question of law which we review de novo.” (City of Oakland v. Oakland Police and Fire Retirement System (2014) 224 Cal.App.4th 210, 228 (City of Oakland).) Before turning to the merits, we begin with a brief overview of res judicata. “The tenets of res judicata prescribe the preclusive effect of a prior final judgment on the merits . . . . Application of the doctrine of res judicata ‘is intended to preserve the integrity of the judicial system, promote judicial economy, and protect litigants from harassment by vexatious litigation.’ [Citation.] It ‘rests upon the sound policy of limiting litigation by preventing a party who has had one fair adversary hearing on an issue from again drawing it into controversy and subjecting the other party to further expense in its reexamination.’ ” (City of Oakland, supra, 224 Cal.App.4th 210, 227–228; accord, Mycogen Corp. v. Monsanto Co. (2002) 28 Cal.4th 888, 896–897; In re Crow (1971) 4 Cal.3d 613, 622–623.) “The doctrine has two distinct aspects: claim preclusion and issue preclusion. [Citation.] Claim preclusion, often referred to as res judicata, provides that ‘a valid, final judgment on the merits precludes parties or their privies from relitigating the same “cause of action” in a subsequent suit.’ ” (City of Oakland, at 6 p. 227; see DKN Holdings LLC v. Faerber (2015) 61 Cal.4th 813, 823–825 (DKN Holdings) [pointing out the terminological confusion that arises from imprecise use of the umbrella term res judicata and explaining the differences between claim preclusion and issue preclusion].) This case involves the second branch of res judicata, which is known interchangeably as issue preclusion and collateral estoppel. Collateral estoppel “precludes relitigation of issues argued and decided in prior proceedings.” (Lucido v. Superior Court (1990) 51 Cal.3d 335, 341 (Lucido), italics added; see DKN Holdings, supra, 61 Cal.4th at p. 824; Vandenberg v. Superior Court (1999) 21 Cal.4th 815, 828– 829 (Vandenberg).) “Traditionally, [courts] have applied the doctrine [of collateral estoppel] only if several threshold requirements are fulfilled. First, the issue sought to be precluded from relitigation must be identical to that decided in a former proceeding. Second, this issue must have been actually litigated in the former proceeding. Third, it must have been necessarily decided in the former proceeding. Fourth, the decision in the former proceeding must be final and on the merits. Finally, the party against whom preclusion is sought must be the same as, or in privity with, the party to the former proceeding. [Citations.] The party asserting collateral estoppel bears the burden of establishing these requirements.” (Lucido, at p. 341.)4 4 “Besides the classic five criteria for applicability, ‘[t]here is an equitable component to collateral estoppel’ as well. (Direct Shopping Network[, LLC v. James (2012) 206 Cal.App.4th 1551,] 1562.) ‘ “[E]ven where the technical requirements are all met, the doctrine is to be applied ‘only where such application comports with fairness and sound public policy.’ ” ’ (Ibid., quoting Smith v. Exxon Mobil Oil Corp. [(2007) 153 Cal.App.4th 1407,] 1414.)” (Union Pacific Railroad Co. v. Santa Fe Pacific Pipelines, Inc. (2014) 231 Cal.App.4th 134, 185.) Thus, courts have crafted policy-based exceptions to collateral estoppel in some circumstances. (See Smith, at p. 1415 [applying “collateral estoppel in this case is unfair and bad policy . . . because of [an] inability at the prior adjudication to obtain expert testimony”]; Wright v. Ripley (1998) 65 Cal.App.4th 1189, 1193–1196 [civil litigation process would “grind to a halt” if orders on motions for sanctions were given collateral estoppel effect because of the incentive lawyers would have to litigate such proceedings exhaustively when faced with the prospect of binding consequences in other proceedings].) Ayala does not invoke any such exception, nor do we perceive a basis for one. 7  It is often said that collateral estoppel “is not an easy rule to apply, for the term ‘issue’ as used in this connection is difficult to define, and the pleadings and proof in each case must be carefully scrutinized to determine whether a particular issue was raised even though some legal theory, argument or ‘matter’ relating to the issue was not expressly mentioned or asserted.” (Clark v. Lesher (1956) 46 Cal.2d 874, 880–881 (Clark).) The identical issue requirement “addresses whether ‘identical factual allegations’ are at stake in the two proceedings.” (Lucido, supra, 51 Cal.3d at p. 342.) “[I]n determining whether the identity [of issues] requirement is satisfied, courts must be mindful of the need to distinguish ‘issues’ from ‘legal theories.’ ” (Wimsatt v. Beverly Hills Weight etc. International, Inc. (1995) 32 Cal.App.4th 1511, 1517 (Wimsatt); Evans v. Celotex Corp. (1987) 194 Cal.App.3d 741, 746 [“Collateral estoppel bars relitigation of the same issues; it does not require identity of legal theories or causes of action.”].) B. The Collateral Estoppel Effect of Unlawful Detainer Judgments Because “[a]n unlawful detainer action is a summary proceeding ordinarily limited to resolution of the question of possession[,] . . . any judgment arising therefrom generally is given limited res judicata effect.” (Malkoskie v. Option One Mortgage Corp. (2010) 188 Cal.App.4th 968, 973 (Malkoskie).) The approach courts take to applying collateral estoppel in this setting is broadly consistent with that taken to the collateral estoppel effect given other forms of summary, informal or specialized adjudication.5 As a general matter in such cases, collateral estoppel will only apply if the party to be bound agreed expressly or impliedly to submit an issue to prior adjudication (see Vandenberg, supra, 21 Cal.4th at p. 835) and had a full and fair opportunity to litigate (see Rohrbasser, 5 See, e.g., Rohrbasser, supra, 179 Cal.App.3d at page 300 (prior findings based solely on affidavits in denial of motion to vacate default and to set aside default judgment); Kelly v. Vons Companies, Inc. (1998) 67 Cal.App.4th 1329, 1335–1337 (prior findings in labor arbitration); Murray v. Alaska Airlines, Inc. (2010) 50 Cal.4th 860, 869 (prior findings in administrative adjudication); Basurto v. Imperial Irrigation District (2012) 211 Cal.App.4th 866, 877–878 (Basurto) (prior findings in employer’s prior internal grievance procedure). 8 supra, 179 Cal.App.3d at p. 300) under circumstances affording due process protections (see Basurto, supra, 211 Cal.App.4th at p. 878). In the specific context we have here, the two leading cases illustrating the collateral estoppel effect of unlawful detainer judgments are Wood v. Herson (1974) 39 Cal.App.3d 737 (Wood) and Vella v. Hudgins (1977) 20 Cal.3d 251 (Vella). In Wood, the defendant (Herson) arranged to obtain a mortgage loan and purchase property allegedly on behalf of a couple who could not qualify for the mortgage (the Woods, one of whom was in bankruptcy). (Wood, at pp. 740–742.) A dispute arose over whether Herson held title and was the true property owner, or, as alleged by the Woods, held title in trust for them and had no right to possession. (Ibid.) Herson brought an unlawful detainer action, and the Woods defended by arguing fraud and claiming that, for a fee, Herson had agreed orally to purchase the property on their behalf and then transfer title to them, but reneged. (Ibid.) They presented this theory through an affirmative defense of fraud. (Ibid.) Herson prevailed, and in a subsequent suit by the Woods for specific performance predicated on the same theory, the trial court granted summary judgment against them, applying the doctrine of collateral estoppel. (Id. at p. 739.) On review, the Court of Appeal affirmed. (Id. at pp. 747–748.) The Wood court’s reasoning, as later described by the California Supreme Court in Vella, was as follows. “Noting that the Woods’ affirmative defense of fraud in the unlawful detainer action was virtually identical to the fraud allegations upon which their suit for specific performance was based, the court concluded that even though title ‘normally is not a permissible issue in an unlawful detainer action,’ the essential issues had been fully and fairly disposed of in the earlier proceeding. [Citation.] The court cited in support of its ruling such varied factors as the length of the ‘summary’ unlawful detainer hearing (seven days), the scope of discovery by the parties (‘extensive’ and ‘complete’), the quality of the evidence (‘detailed’), and the general character of the action (‘[clearly] . . . not the customary unlawful detainer proceeding’). [Citation.] A lengthy and comprehensive superior court record replete with precise findings of fact persuaded the Wood court that application of collateral estoppel to curtail further 9 litigation would involve ‘no miscarriage of justice—(the) Woods have had their day in court . . . .’ ” (Vella, supra, 20 Cal.3d at p. 256, quoting Wood, supra, 39 Cal.App.3d at pp. 740, 742, 745.)6 The Supreme Court adopted the same frame of analysis but reached the opposite conclusion on the facts in Vella, a case in which a property owner in an eviction action tried to block subsequent litigation against him by the evicted tenant based on the unlawful detainer judgment he obtained. There, the plaintiff, Nancy Vella, alleged she had a long-term “confidential” and “intimate” relationship with the defendant, Everett Hudgins. (Vella, supra, 20 Cal.3d at p. 253.) When Vella had trouble paying a note on a second deed of trust on her home, Hudgins purchased the note, told Vella he was doing so to protect her from default, and assured her she need not worry about making payments on the obligation. (Id. at p. 254.) The two then quarreled; Hudgins directed the deed of trust trustee to give notice of default; and in an ensuing unlawful detainer proceeding by Hudgins against Vella, Vella pleaded fraud as an affirmative defense, but ultimately lost and was evicted. (Ibid.) Nonetheless, she managed to persuade the trial court to allow her to litigate a fraud claim against Hudgins in a subsequent action, and she prevailed in a jury trial. (Ibid.) On appeal, the Supreme Court rejected Hudgins’s contention that Vella’s separate fraud action was barred by res judicata. (Ibid.) The Supreme Court distinguished Wood, describing it as an “uncommon” situation in which the parties to an unlawful detainer action had fully and extensively litigated issues beyond simple possession. (Vella, supra, 20 Cal.3d at p. 257.) The court acknowledged the Wood holding that “ ‘full and fair’ litigation of an affirmative defense—even one not ordinarily cognizable in unlawful detainer, if it is raised without objection, and if a fair opportunity to litigate is provided—will result in a judgment conclusive upon issues material to that defense” (Vella, at pp. 256–257), but explained 6 See also Gombiner v. Swartz (2008) 167 Cal.App.4th 1365, 1371 (ruling in unlawful detainer proceeding that rent stabilization ordinance applied, thus justifying withholding of rent as an offset against past overpayment of excess rent, given collateral estoppel effect in breach of lease case by landlord for nonpayment of rent). 10 that the facts presented in Vella were more typical of a summary unlawful detainer proceeding than the situation presented in Wood. “The record offered in support of the plea of res judicata is virtually barren. Evidently the unlawful detainer proceedings were unrecorded or untranscribed, for no transcript of the municipal court hearing exists, and no findings of fact or conclusions of law were made, other than a notation in the trial judge’s minute order to the effect that Vella had not proved her affirmative defenses of ‘waiver and [equitable] estoppel and tender.’ The sparse record presented to us fails to show either the precise nature of the factual issues litigated, or the depth of the court’s inquiry. We decline to assume, given the summary character of this type of action, that the mere pleading of a defense without objection by the adverse party necessarily demonstrates adequate opportunity to litigate the defense. The fact that in the unlawful detainer action both parties submitted trial-length estimates of two hours, whereas trial of the second action consumed four days, while not controlling, does create a strong inference that the former proceeding was a conventional unlawful detainer action, unlike the elaborate and highly atypical proceeding considered in Wood.” (Vella, at p. 258.)7 C. Application of Collateral Estoppel in this Case This case, in our view, is more akin to Wood than to Vella. Ayala could have moved to consolidate the unlawful detainer proceeding with this action, thus requiring the court to determine whether the issues presented were so complex and so intertwined with the issue of title that “ ‘the entire case [should be] treated as an ordinary civil action, not as a summary proceeding’ ” (Martin-Bragg v. Moore (2013) 219 Cal.App.4th 367, 387), but he did not do so. Instead, he “acceded to the summary and expedited procedures of unlawful detainer with respect to” his claim to equitable title. (Id. at p. 389.) By 7 See also Gonzales v. Gem Properties, Inc. (1974) 37 Cal.App.3d 1029, 1032, 1036 (unlawful detainer judgment against homeowner who suffered foreclosure and was evicted by holder of second mortgage does not bar later action against note holder and those who acted in concert with him, individuals who “were widely known in their trade as persons willing to employ tricks and devices to acquire property for amounts below its actual value,” based on alleged scheme to acquire title fraudulently by manipulation of trustee sale). 11 recharacterizing the governing contract as an oral installment sale contract for the purchase of land, rather than a lease, Ayala hoped to persuade Judge Daniels he was not a tenant subject to eviction. The premise of Ayala’s attack on jurisdiction in the unlawful detainer proceeding was that Dawson, in breach of his fiduciary duty as a real estate broker, duped Ayala into signing the lease, while misrepresenting it to be an installment sale contract. Judge Daniels considered this issue after holding an evidentiary hearing that ran six and a half courtroom hours, taking testimony from both parties, and admitting extensive documentary evidence. In the end, he specifically rejected the notion that Dawson was Ayala’s real estate broker or that Ayala should be excused from performance under the lease for lack of awareness of its terms. In reviewing the later decision by Judge Kays to give Judge Daniels’s rejection of Ayala’s fraud-in-the-inducement argument collateral estoppel effect, we determine what was “actually litigated” by looking to the pleadings and the proof. (Clark, supra, 46 Cal.2d at pp. 880–881.) “An issue is actually litigated ‘[w]hen [it] is properly raised, by the pleadings or otherwise, and is submitted for determination, and is determined . . . .’ ” (People v. Sims (1982) 32 Cal.3d 468, 484, original italics, superseded by statute on other grounds as noted in Gikas v. Zolin (1993) 6 Cal.4th 841, 851.) There was far more here than what the Supreme Court characterized in Vella as the “mere pleading of a defense without objection . . . .” (Vella, supra, 20 Cal.3d at p. 258.) The record in the unlawful detainer proceeding is clear that Ayala elected to place his fraud theory in issue by moving to quash on the ground there was no valid lease; that he framed the issue by persuading Judge Daniels to take judicial notice of his verified allegations of fraud in this case; that this fraud allegation was the central focus of an evidentiary contest in a long-cause hearing at which both Ayala and Dawson testified; that Ayala’s presentation at the hearing was fulsome and unrestricted; and that Judge Daniels ultimately decided the fraud issue against him. Prior to the hearing on the motion to quash, Ayala was given the opportunity to take discovery, and he in fact took Dawson’s deposition, which he then used extensively, 12 but ultimately to no effect, in cross-examining Dawson at the hearing. Unlike the situation in Vella, there was an extensive record, consisting of a 243-page transcript and 21 documentary exhibits, among them the written lease signed by Dawson and Ayala, the ledger of rent payments made by Ayala, and the amortization and payment records for Dawson’s mortgage loans on the property. Also included is Judge Daniels’s on-the- record explanation of his findings. Judge Daniels found, much as the trial judge in the Wood case did, that this was an unusual case. “[M]y goodness,” he stated before announcing his findings, “this is certainly not a typical transaction.” After observing that the transaction seemed designed to be a means of financing for Ayala—which is what Dawson had argued, and is what a lease-purchase option accomplishes, where the option is exercised—he proceeded to characterize the parties’ relationship legally as one of landlord-tenant. Ayala had the right to seek review of that determination, and availed himself of this right by pursuing a petition for a writ of mandate in the superior court appellate division, without success.8 8 The appellate division issued a reasoned order denying the petition after undertaking a thorough review of the record, concluding as follows: “It is evident from the trial court’s recitation of reasons for its decision that the decision is supported by substantial, if not overwhelming, evidence. [¶] Petitioner admitted that he executed a written lease agreement that gave him an option to purchase the property . . . . The agreement was clearly labeled ‘RESIDENTIAL LEASE WITH OPTION TO PURCHASE (CALIFORNIA SHORT FORM)’, established that Petitioner would pay monthly rent of $2,700, indicated that Petitioner would pay nonrefundable option consideration of $77,000, and specifically provided that Petitioner could make any legally conforming use of the property, that Petitioner was solely responsible for maintenance of the property, and that Petitioner was free to sublease without Respondent’s consent . . . . Each page was initialed, confirming that Petitioner had read the page, and Petitioner’s signature established that Petitioner had ‘thoroughly read and approved’ every provision of the agreement . . . . [¶] Notwithstanding Petitioner’s claims that his primary language was Spanish and that he did not actually read the document prior to initialing and signing it, his own testimony established that he was familiar with legal requirements for written contracts as part of his occupation as a licensed contractor, and that he attended and graduated from high school in the United States . . . . [¶] The terms of the agreement, no matter what Petitioner may or may not have believed, were consistent with the terms orally discussed between the parties. Respondent would purchase the property in Respondent’s name . . . . Petitioner would 13  In his writ petition, Ayala argued, as he does here, that Judge Daniels decided only the issue of possession and that the separate and distinct issue of title is not germane in an unlawful detainer proceeding. In the typical case, that is true. But what the argument overlooks is the necessary factual predicate for Judge Daniels’s ultimate conclusions—he found that Dawson carried his burden of proof on the question of jurisdiction, that there was an enforceable written lease agreement, and thus, impliedly, that Dawson did not commit fraud—fundamentally undercuts Ayala’s claim to equitable title. Indeed, this is exactly what the trial judge in Wood, supra, 39 Cal.App.3d 737 found, where a couple who could not qualify for a loan used a front buyer and then failed when they tried to argue that the strawman financier they relied upon was not the true owner. Just as that couple was barred from litigating such a defense twice, Ayala was barred here. Judge Kays was correct to so rule. In a final attempt to avoid the collateral estoppel effect of Judge Daniels’s adverse ruling against him in the unlawful detainer proceeding, Ayala advances two additional arguments. He contends, first, that because the final order in the unlawful detainer proceeding was a “clerk’s judgment” of default, there was no decision on the merits, and, second, even if there was a decision on the merits, because the appellate division panel expressly stated in denying his writ petition that the unlawful detainer proceeding was strictly limited to the issue of possession, the denial of the writ deprives Judge Daniels’s findings of any collateral estoppel effect here under the “law of the case” doctrine. We see no merit in either point. Ayala appeared specially in the unlawful detainer proceeding, contested the issue of jurisdiction—tendering his fraud-in-the-inducement theory as the linchpin issue—was provide the down payment necessary to purchase the property . . . . Petitioner would pay Respondent an amount equivalent to the monthly mortgage, interest, insurance and taxes owed on the property plus $200 . . . . The only real dispute is whether the agreement was to be structured as a lease with an option to purchase or an installment land sales agreement, an issue that was never orally discussed prior to the execution of the lease agreement.” (Denial of Petition for Writ of Mandate, November 6, 2012, Writ No. FCS 040463 at pp. 3–4, record citations omitted.) 14 given an evidentiary hearing, and lost on the merits. He does not and cannot argue that Judge Daniels’s order was somehow tentative, subject to change, or never reached a point of finality. For purposes of res judicata, it is deemed to be part of the judgment. (McClain v. Rush (1989) 216 Cal.App.3d 18, 28 [“for purposes of issue preclusion . . . ‘final judgment’ includes any prior adjudication of an issue in another action that is determined to be sufficiently firm to be accorded conclusive effect”], quoting Rest.2d Judgments, § 13, italics omitted.) As to Ayala’s “law of the case” argument, we do not view the appellate division’s observation that the unlawful detainer proceeding was strictly limited to the issue of possession as having any bearing here. The passage from its opinion that Ayala relies upon addresses due process, which is relevant to collateral estoppel but is not the same thing. And in any event, even if the appellate division panel had intended to address the collateral estoppel effect of Judge Daniels’s order, its determination on the point is ultimately no more binding on us than a finding by Judge Daniels on the collateral estoppel effect of his own order. (See Velasquez v. Superior Court (2014) 227 Cal.App.4th 1471, 1477, fn. 7 [“Appellate division decisions have persuasive value, but they are of debatable strength as precedents and are not binding on higher reviewing courts.”].) D. Appeal of the Attorney Fees Award Ayala’s appeal of the attorney fees award against him rises or falls with his appeal of the judgment. He contends only that “[i]f a judgment is reversed and there is also a related appeal from an order awarding attorney fees and/or costs based on the judgment, the attorney’s fees and/or costs award will be reversed.” (Citing Purdy v. Johnson (1929) 100 Cal.App. 416, 420–421.) Since we affirm the judgment, we affirm the award of attorney fees as well. III. CONCLUSION AND DISPOSITION The judgment is affirmed and the order awarding attorney fees is affirmed. Dawson shall recover his costs. 15  _________________________ Streeter, J. We concur: _________________________ Ruvolo, P.J. _________________________ Rivera, J. 16 A142830/Ayala v. Dawson A142830 – Ayala v. Dawson Trial Court: Solano County Superior Court Trial Judge: Hon. Scott L. Kays Counsel: Arthur Samuel Humphrey, Attorney at Law and Arthur Samuel Humphrey for Plaintiff and Appellant. Terry A. Duree for Defendant and Respondent. 17 

On Thursday, June 13, 2013, Yitzchak Scott-Thoennes wrote: >> I think it may be easier to list the cases that could be expected to > *continue* to work in the future; It's my understanding that smart match will continue to work *as* *is* in all future versions of Perl if you enable experimental smart match behaviour in your code. The point in the declaration is to say 'I want you to try to have the behaviour that existed when this was an experimental feature, not whatever the behaviour is now' - i.e. to tell whatever version of Perl your code is running on to always use smart match as it worked in 5.18. The *warning* in 5.18 is to tell you that you're *not* declaring you want the current experimental behaviour (i.e. behaviour in 5.18 and previous versions of Perl) but whatever behaviour some crazy person on p5p implements in some yet to be written Perl. Turning off the warnings (with 'no warnings') rather than using the experimental feature (with 'use') is saying 'I don't want the code to work as is but break when someone changes something'. Hth Mark

--- abstract: 'It is commonly assumed in percolation theories that at most one percolating cluster can exist in a network. We introduce sausage-like networks (SLNs), an ensemble of synthetic modular networks in which more than one percolating cluster can appear. We show that coexisting percolating clusters (CPCs) emerge in such networks due to limited mixing, i.e., a small number of interlinks between pairs of modules. We develop an approach called modular message passing (MMP) to describe and verify these observations. We demonstrate that the appearance of CPCs is an important source of inaccuracy in the previously introduced percolation theories, such as the message passing (MP) approach. Moreover, we show that the MMP theory improves significantly over the predictions of MP for percolation on synthetic networks with limited mixing and also on several real-world networks. These findings have important implications for understanding the robustness of networks and in quantifying epidemic outbreaks in the susceptible-infected-recovered (SIR) model of disease spread.' author: - Ali Faqeeh - Sergey Melnik - 'Pol Colomer[-]{}de[-]{}Sim[ó]{}n' - 'James P. Gleeson' bibliography: - 'networks.bib' title: Emergence of coexisting percolating clusters in networks --- Percolation theories are among the most studied in network science [@Newman10], as well as in several other areas [@christensen05; @stauffer1994], providing insights for a broad range of applications such as robustness of a network to random failures or attacks [@Callaway00], epidemics in contact processes [@Munoz10_PRL], vaccination strategies [@Newman10], neuronal avalanches [@Friedman_choas13], and stability of gene regulatory networks [@Squires_prl12]. In the simplest case of bond (or site) percolation, a fraction $p$ of the links (nodes) are randomly chosen to be occupied and the rest of the links (nodes) are removed from the network [@Newman10]. The quantity of interest is $\mathit{\overline{S}}$: the expected fractional size of the giant component (GC) of the network, which in the limit of infinitely large networks is referred to as the percolating cluster (PC) of the network [@Newman01e; @christensen05]. The size of the GC scales linearly with the network size while the fractional sizes of other clusters vanish in the limit of infinitely large networks. Theoretical approaches and extensive numerical simulations play pivotal roles in understanding and describing the behavior of percolation processes on networks. The $p_k$ theory for bond percolation [@Melnik11; @Newman10], for example, can accurately describe the results of numerical simulations on configuration model [@Molloy95] networks using only the network degree distribution. On networks with degree-degree correlations, the accurate results are obtained using the so called $P(k,k')$ theory [@Vazquez03] which employs the joint degree distribution. The $P_{k,k'}^{i,i'}$ theory [@Melnik_chaos14] can provide a more accurate description of dynamics on modular networks, as it considers the joint degree distributions within and between modules. The message passing (MP) approach [@karrer_PRL14] provides more accurate results than the aforementioned theories as it uses the full information on the adjacency of individual nodes, and reduces to the above degree-based approximations in special cases [@Faqeeh15]. As these theories assume the network is locally treelike, they are prone to errors in clustered networks which have an appreciable density of short loops [@Melnik11; @Faqeeh15]. However, on some real-world clustered networks these theories still perform well, and in some other cases the inaccurate predictions of these theories are shown to be only partly caused by the presence of short loops [@Faqeeh15]. This indicates the presence of an unexplained source of error and possibly a phenomenon not captured by the theories. In this paper, we show the appearance of coexisting percolating clusters (CPCs) in certain networks, and demonstrate that this phenomenon causes significant errors in the aforementioned theories. We show that CPCs appear in modular networks with limited mixing, i.e., networks with a sufficiently small (limited) number of interlinks between modules. We verify these observations by developing the modular message passing (MMP) theory which takes into account the presence of independent CPCs. We show that the MMP theory provides accurate predictions on treelike modular networks with limited mixing and also improves over the predictions of MP on several real-world clustered networks. We begin by introducing sausage-like networks (SLNs), a simple ensemble of random networks that demonstrates the appearance of CPCs. To create an SLN we first pick a graph with size $N_m$, which can be any connected undirected unweighted graph. Then we make $M$ identical copies of that graph (Fig. \[f1a\]) which will become modules in the SLN. We assign to each of these modules a unique label $m\in\{1,2,\ldots,M\}$ and connect each pair of modules with consecutive labels $m$ and $m+1$ by $I$ links. To do so, exactly $I/2$ links are selected randomly from module $m$. For each selected link $i_m\mbox{---}j_m$ we consider its copy $i_{m+1}\mbox{---}j_{m+1}$ in module $m+1$, and rewire these two links to create two new links $i_{m}\mbox{---}j_{m+1}$ and $i_{m+1}\mbox{---}j_{m}$ instead. The resulting SLN is comprised of a chain of modules (Fig. \[f1b\]), each pair of consecutive modules connected with exactly $I$ interlinks. Moreover, an SLN has a degree distribution and degree-degree correlations between and beyond the nearest neighbors identical to those of the original graph [@Faqeeh15]. Similarly, one constructs SLNs from non-identical modules by rewiring links that are not copies but randomly selected from each module. \ In Figs. \[f1c\]-\[f1e\], we illustrate the bond percolation results on SLNs constructed from a 3-regular graph. The results from the MP theory are obtained using the Eqs. (1) and (2) of Ref. [@Faqeeh15] (originally due to [@karrer_PRL14]): $$\begin{aligned} u_{ij}=1&-&p+p\prod_{k\neq i}{A}_{jk}u_{jk},\\ \mathit{\overline{S}}=\frac{1}{N}\sum_{i=1}^Ns_i&=&1-\frac{1}{N}\sum_{i=1}^N\prod_j{A}_{ij}u_{ij},\label{eq0b}\end{aligned}$$ where $p$ is the occupation probability, $u_{ij}$ is the probability that node $i$ is not connected to the network PC via its link to $j$, $A$ is the network adjacency matrix, $N$ is the size of the network, and $s_i=1-\prod_j{A}_{ij}u_{ij}$ is the probability that node $i$ is in the PC. Numerical simulations show that the behavior of $\mathit{\overline{S}}$ in the SLNs depends substantially on the number of interlinks $I$ and the number of modules $M$. Surprisingly, the result of the MP theory for the SLNs is independent of $I$ and $M$ and coincides with its prediction for a single 3-regular graph (Fig. \[f1c\]); it is worth mentioning that in such SLNs the result of MP theory is the same as the results of any of its degree-based reductions mentioned above. The numerical results, on the other hand, deviate from the theoretical prediction as the number of interlinks $I$ is decreased (Fig. \[f1c\]). We quantify the difference between the theoretical and numerical results by calculating the mean absolute error between the two: $E=1/R\sum_{j=1}^R|\mathit{\overline{S}}_{theo}(p_j)-\mathit{\overline{S}}_{num}(p_j)|$, where the sum is over $R=100$ equally spaced occupation probabilities $p_j=j/R$. Figure \[f1d\] shows that the error increases dramatically for lower number of interlinks $I$; on the other hand the error increases only slightly for smaller module sizes $N_m$ (Fig. \[f1e\]). Figure \[f1e\] highlights also that the error increases for lower $I$ rather than for lower ratio of $I$ to all edges. To understand these observations, it is necessary to inspect also the numerical results for $S$, the fractional size of the network PC for single realizations of the bond percolation process. Figure \[f2a\] shows the results for a 3-regular SLN with $M=5$ modules; the MP theory overestimates $\mathit{\overline{S}}$, which is the expected (average) value of $S$. For a sufficiently large network, we normally expect that $S$ fluctuates slightly around $\mathit{\overline{S}}$. Surprisingly, we observe in Fig. \[f2a\] that, for a fixed value of $p$ (in a certain interval), $S$ can take one of several possible values which can be significantly different from $\mathit{\overline{S}}$. The different possible values for $S$ can be explained as follows. Let us denote by $S_m$ the fraction of nodes in module $m$ that are in the network PC predicted by the theory. Hence, for the MP theory $S_m=\frac{1}{N}\sum_{i\in\hat{m}}s_i$, where $\hat{m}$ denotes the set of nodes located in module $m$, and Eq.  becomes: $$\label{eq0c} \mathit{\overline{S}}=\sum_{m=1}^MS_m.$$ In Fig. \[f2a\], we can observe that $\mathit{\overline{S}}$ predicted by MP matches the largest possible value of $S$. The lowest possible value of $S$, on the other hand, coincides with $S_m$ of only one module. The next 3 larger possible values of $S$ coincide with the sum of $S_m$ for 2, 3, and 4 modules respectively (Fig. \[f2a\]). Consider another example of an SLN consisting of one 3-regular module and one 4-regular module each having the same number of nodes $N_m$, and let us denote by $S_1$ and $S_2$ the $S_m$ of the 4-regular and 3-regular graph respectively. For this SLN, $\mathit{\overline{S}}$ from numerics and from the MP theory both match $S_1$ up to $p_{3reg}=0.5$, the percolation threshold of a 3-regular graph. Above this value, the MP prediction deviates from the numerical result (Fig. \[f2b\]). This deviation can be better understood by looking at single realizations of the Newman-Ziff algorithm [@Newman01e], where starting with no occupied links, we occupy links one by one in random order. As the network is large, $p$ is approximately equal to the fraction of occupied links. In Fig. \[f2b\], we can observe that up to $p_{3reg}$, single realization values of $S$ match the value of $\mathit{\overline{S}}$. However, above $p_{3reg}$, while the predicted $\mathit{\overline{S}}$ is $S_1+S_2$, $S$ will remain equal to $S_1$ until some larger value of $p$ and then suddenly jump to $S_1+S_2$. \ This implies that although $S_2$ is finite for $p\geqslant p_{3reg}$, it does not represent the probability a node in module 2 belongs to the network PC, and accordingly $S_1+S_2$ is not the expected size $\mathit{\overline{S}}$ of the network PC. In fact, the nodes in module 2 are part of a PC with size $S_2$ and those in module 1 are part of another PC with size $S_1$. In a single realization these two PCs may be unconnected. Then, when more links are occupied one at a time, at a value of $p$ with $p\geqslant p_{3reg}$ the two PCs become suddenly connected, and the size of the network PC changes abruptly from $S_1$ to $S_1+S_2$. We refer to such PCs, which exist independently of each other in the network but may only be connected with a finite probability, as coexisting percolating clusters (CPCs). Next, we develop the modular message passing (MMP) theory to describe and verify the phenomenon of CPCs appearing in networks. The two main assumptions are (i) modules can percolate independently (hence, the appearance of independent CPCs), and (ii) PCs of neighboring modules are connected with probability $\eta<1$. This is a new concept different from the common assumption (see for example [@Newman10; @Melnik_chaos14; @colomer2014double]) that there exists only one monolithic PC in a network; here we show that in networks with limited mixing, the network PC is polylithic, i.e., constituted by CPCs that are connected together. In such networks, the CPCs emerge independently inside the internally well connected modules, but they still may not be connected to each other due to the small number of interlinks between the modules. For $p<1$, interlinks may be unoccupied; the CPCs of two neighboring modules are connected if and only if they share at least one occupied interlink that connects nodes from the CPCs of the two modules. For SLNs we assume that each boundary node (a node with links to other modules) has exactly one link to a neighboring module. Then for an SLN with two modules, our MMP theory is comprised of two simple equations. First we calculate $\eta_{mn}$, the probability that the CPCs of the two modules $m$ and $n$ are connected: $$\label{eq1}\centering \eta_{mn}=1-\left(1-pv_mv_n\right)^I.$$ Here $p$ is the occupation probability and $v_m$ is the probability that a boundary node in $m$ is part of the CPC of $m$. Then $1-pv_mv_n$ is the probability that the two CPCs are not connected via an interlink; raising this term to the power of $I$ gives the probability that they are not connected via any of the $I$ interlinks. Equation  is independent of $N_m$, hence in the thermodynamic limit ($N\to\infty$), if $I$ is fixed to a sufficiently small number, (independent) CPCs emerge, connected with a probability $\eta_{mn}(p)<1$. Whereas for large $I$, since $\eta_{mn}\to 1$ the two CPCs are connected with high probability, leading to a monolithic PC. For the SLNs described above, $v_m$ and $v_n$ can be simply calculated using the $p_k$ theory (see Sec. S.1. of the Supplemental Material (S.M.) [@SM]). For an SLN with two modules, the size of network PC is then: $$\label{eq2} \mathit{\overline{S}}=\eta_{12}\left(S_1+S_2\right)+\left(1-\eta_{12}\right)S_1,$$ where $S_1$ and $S_2$ are the fractional sizes of respectively the larger and the smaller CPC of the SLN. Figure \[f2c\] shows that the prediction of the MMP theory (Eqs. -) matches perfectly the numerical result for the SLN of Fig. \[f2b\]. For SLNs with more than two modules Eq.  can still be used to calculate $\eta_{mn}$ for each pair of modules $m$ and $n$. In the case when boundary nodes have more than one interlink or when the connection pattern of the modules can not be well approximated using the $p_k$ theory, Eq.  should be extended to include more information on the network structure. We can use the full information on the adjacency of individual nodes to write a general formula for the connection probabilities $\eta_{mn}$ between CPCs in a treelike network: $$\label{eq3} \eta_{mn}=1-\prod_{i=1}^I\left[1-\left(1-\prod_{j\in\mathcal{A}_m(i)}u_{ij}\right)\left(1-\prod_{k\in\mathcal{A}_n(i)}u_{ik}\right)\right],$$ where $i$ is a boundary node of module $m$ and $\mathcal{A}_m(i)$ denotes the set of neighbors of $i$ in modules $m$. Here, $u_{ij}$ and $u_{ik}$ are the probabilities that $i$ is not connected to the CPC of, respectively, modules $m$ and $n$ via its links to nodes inside each of those modules. Hence, within the first (second) set of parentheses in Eq.  is the probability that $i$ is in the CPC of $m$ (the CPC of $n$), and in the square brackets we have the probability that the two CPCs are connected via the interlinks of $i$. Therefore, $\eta_{mn}$ is the probability that the two CPCs are connected via any of their $I$ interlinks. To calculate $\mathit{\overline{S}}$ for networks with more than two modules (and consequently more than two CPCs), Eq.  should be extended as well. If a networks contains several CPCs then in a single realization of percolation, different CPCs (with sizes $S_m$) are connected together with some probability, creating larger polylithic PCs. The polylithic PC $l$ has a size $S^{\rm(pol)}_{l}=\sum_{m\in\hat{l}}S_m$, where $\hat{l}$ denotes the set of CPCs that constitute $l$. Then in a single realization at a fixed value of $p$, the size $S$ of the network PC is $\max_l\left(S^{\rm(pol)}_{l}\right)$, i.e., the size of the largest polylithic PC. Then the expected size of the network PC is $$\label{eq4} \mathit{\overline{S}}=\sum_{S}P_{\left(S\right)}{S},$$ where $P_{\left({S}\right)}$ is the probability that in a single realization the size of the largest polylithic PC is ${S}$. To calculate $P_{\left({S}\right)}$, for each $p$ we first calculate $S_m(p)$ values using the MP theory or an appropriate degree-based reduction of MP. Then we assume a meta-network which is comprised of meta-nodes; each meta-node $m$ represents a module of the original network and has a weight $S_m(p)$. In each realization of the meta-network for a fixed value of $p$, each pair of meta-nodes $m$ and $n$ are connected with probability $\eta_{mn}(p)$. For sufficiently large number of meta-network realizations, we calculate $P_{\left({S}\right)}$ using the Newman-Ziff algorithm [@Newman01e] with the following modifications: (i) each link is added with probability $\eta_{mn}(p)$, and (ii) the size of a cluster $l$, comprised of connected meta-nodes, is the sum of the weights of meta-nodes it includes, i.e., $S^{\rm(pol)}_{l}$. These calculations are performed very quickly, as the number of the modules is usually much smaller than the number of nodes. Figure \[f2d\] illustrates that, using Eqs. -, the MMP theory performs very well for the SLN of Fig. \[f2a\]. As shown in Figs. \[f2a\] and \[f2b\] in the presence of CPCs the values of $S$ for single realizations can deviate considerably from the expected value $\mathit{\overline{S}}$. In the MMP theory, the variability of $S$ is determined by $P_{(S)}$ defined above. The percentiles calculated from $P_{(S)}$ match very well the numerical values in SLNs (see Fig. \[f2d\] for example), which confirms that the high variability of $S$ originates from the presence of CPCs that suddenly merge together. Similar results (not shown) are obtained for SLNs with different values of $I$, $M$, $N_m$, and also for SLNs constructed from modules with a heterogenous structure, i.e., with a power-law degree distribution. We also provide results for LFR benchmark networks [@Lancichinetti08], as an example of modular networks with heterogeneous structure. In LFR networks, all pairs of modules can be connected without any restriction (as apposed to SLNs), and the node degrees and community (module) sizes have a power-law distribution. On LFR networks with low mixing between modules, the numerical results do not match the MP predictions, whereas MMP method provides accurate predictions (e.g., see Fig. \[f4a\]). As mentioned before, MMP assumes that modules have (independent) CPCs. This assumption holds only for sufficiently large modules. Nevertheless, if the size of the modules is as small as $N_m^{\rm(min)}\simeq50$, MMP still performs well (see S.M., Sec. S.2. [@SM]). We can show that on several real-world networks the bond percolation results are affected by the emergence of CPCs. To do so, we first identify the best representations of network modular structure using a multiresolution community detection method (e.g., see Refs. [@Ronhovde09; @blondel08; @Faqeeh12; @Rosvall08]). Then, we choose the representation that maximises the modularity $Q$ [@Newman04_pre] and minimizes $M_s$ the number of modules with sizes smaller than $N_m^{\rm(min)}=50$, i.e., the representation with maximum $\frac{(M-M_s)}{M}Q$. Figure \[f4b\] shows that MMP improves significantly over the MP prediction of $\mathit{\overline{S}}$ on the western United States power grid network [@Melnik11; @Watts98], and also provides a prediction for the variability of $S$ according to the percentiles of $P_{(S)}$. The results of MMP on the power grid network, match the numerics for the $P_{k,k'}^{i,i'}$ rewired  [@Melnik_chaos14] version of this network in which the links are rewired inside each module; $P_{k,k'}^{i,i'}$ rewiring preserves the modular structure and degree-degree correlations but effectively destroys the short loops. This shows that MMP provides a highly accurate prediction in the absence of short loops and when the modular structure is identified accurately. Similar results are shown in S.M. for several other examples of real-world networks [@SM]. In summary, we demonstrated that CPCs can emerge in ensembles of random networks and in real-world networks, when the network modules are connected via a small number of interlinks. Moreover, we showed that CPCs are an important source of error in the theories for bond percolation and proposed the MMP theory that accurately captures the impact of CPCs on percolation results. An important implication of the appearance of CPCs is the uncertainty they cause in determining the network robustness: when the CPCs emerge, the size ${S}$ of the network PC can be highly variable (as shown by the percentiles). This implies the prominent role of interlinks in network robustness, even in the absence of module-based targeted attacks [@daCunha15], and subject to only random failures. Another implication is that the eventual size of an epidemic spread in the SIR model [@Newman02b; @Karrer10] may not be best represented by $\mathit{\overline{S}}$, which is the expected size of the largest polylithic PC, since even CPCs not in the largest polylithic PC represent (independent) outbreaks of comparable sizes located in different modules. Hence, the total size of an epidemic outbreak may better be represented by the sum of the sizes of all CPCs. We would like to thank Marián Boguñá, Mel Devine, and David O’Sullivan for valuable comments. We acknowledge funding from Science Foundation Ireland (under programmes 11/PI/1026 and 12/IA/1683) and from the European Commission FET-Proactive project PLEXMATH (FP7-ICT-2011-8; Grant No. 317614).

Thursday, November 29, 2012 Which is the Dumbest Professional Sports League? Today's question comes after learning that the NHL and NHLPA failed to make progress with federal mediators. No new meetings are scheduled with the mediators and both sides don't plan on using mediation going forward. So here's the question; which is the dumbest professional sports league in North America? I'm not going to even list other leagues. There isn't any league that comes close to the NHL when using the word dumb. The NHL wins hands down. Congratulations to the NHL and NHLPA for becoming the Rome of sports. Actually, they were never Rome. They've never been on top. Below are the reasons why I think that the NHL is the dumbest league 1. Threw away any chance of capitalizing on a record year Only dummies would throw away a chance of capitalizing on a record season. The 2011/12 season was the most successful in NHL history. So what was the NHL's solution to building on that year? Lockout! Lockout! Lockout! What more can be said on this subject. 2. Would rather have an extended vacation than make money It seems like both the NHL and NHLPA would rather take an extended vacation than earn some money. With the inevitability of the 2012/13 season being cancelled the NHL will lose billions of dollars. They would rather break the backs of the players. The players will lose hundreds of millions of dollars. These stubborn mules would rather give up their earnings than negotiate in good faith. They also want to win at all costs. I believe that if the season is cancelled, the players will end up losing the battle as far as earnings are concerned even if they win the negotiations war. That's dumb. 3. Failed mediation Federal mediators had reached out to the NHL and NHLPA to help them figure out a solution to this mess. In just two sessions and no progress both sides say that the mediation route is over. This is further proof that these guys aren't serious about getting a deal done. As mentioned above they like their vacations too much. 4. Ranked fourth out four major professional sports leagues This isn't really a reason but a fact. The NHL is the lowest ranked major sports league in North America. The NHL can't afford to lose their fan base. Their fan base is much less than the other leagues and less loyal (except for Canada). I think it's silly and foolish for the NHL to have gone down this road. The NHL has proven to me that is the dumbest sports league on this continent and is made up of dummies. As a fan of the NHL I have lost all faith in the people that run the league and play in the league.

Q: View Pager with OnLongClickListener doesn't work I have a ViewPager on which I am trying to apply onLongClickListener to change a Fragment from Activity which does not work. mPager1.setOnLongClickListener(new View.OnLongClickListener() { @Override public boolean onLongClick(View v) { onViewPagerClick(); return false; } }); Fragment replacement: public void onViewPagerClick() { FragmentTransaction ft5 = getSupportFragmentManager().beginTransaction(); fragment = new SingleProductFragment(); container.setVisibility(View.VISIBLE); ft5.add(R.id.container, fragment); // ft.addToBackStack(null); ft5.commit(); sv.setVisibility(View.INVISIBLE); addButton.setVisibility(View.INVISIBLE); } A: ViewPager doesn't consume click events so try to set the listener on your image or container layout instead of viewPager in instantiateItem(). UPDATE @Override public Object instantiateItem(ViewGroup container, int position) { LayoutInflater inflater = LayoutInflater.from(MainActivity.this); ViewGroup viewPagerItem = (ViewGroup)inflater.inflate(R.layout.viewPager_item, container, false); ImageView imageView = (ImageView)viewPagerItem.findViewById(R.id.imageView); viewPagerItem.setOnLongClickListener(new View.OnLongClickListener() { @Override public boolean onLongClick(View view) { Toast.makeText(MainActivity.this, "TEST", Toast.LENGTH_LONG).show(); return false; } }); container.addView(viewPagerItem); return viewPagerItem; }

Introduction {#s1} ============ Tuberculosis (TB) remains a leading human infectious disease and a major public health problem in low-income countries [@pone.0011019-World1]. Despite the availability of the Bacillus Calmette-Guerin (BCG) vaccine for more than 80 years, until now an effective tuberculosis vaccine is still far to be generated and still unknown are the correlates of protection against this disease [@pone.0011019-Barker1]. There is a remarkable body of evidence that IFN-γ producing CD4^+^ T cells are the main mediators of protection against *Mycobacterium tuberculosis* (MTB) infection both under natural conditions and following vaccination [@pone.0011019-Cooper1], [@pone.0011019-Jung1]. However, while activated CD4^+^ T cells are required to avoid the spread of MTB during infection, they can also cause severe inflammation with collateral tissue damage when not tightly controlled [@pone.0011019-Dannenberg1], [@pone.0011019-GonzalesJuarrero1]. Granuloma necrosis in a mouse model of MTB-induced pulmonary immunopathology is due to IFN-γ and T cells expressing the αβ T cell receptor [@pone.0011019-Ehlers1]. Moreover IFN-γ-producing CD4^+^ T cells can interfere with development of protective immunity during experimental vaccination with mycobacterial antigens [@pone.0011019-Elias1]--[@pone.0011019-Palma1], including Ag85B, an abundant secreted protein of replicating MTB which is currently evaluated in various TB vaccine formulations [@pone.0011019-Barker1], [@pone.0011019-Dietrich1]--[@pone.0011019-Hoft1]. It has been recognized that activation of Ag85B-specifc CD4^+^ T cells in TB patients is not always associated with a favourable prognosis [@pone.0011019-Boesen1], [@pone.0011019-Schwander1], and that frequencies of Ag85-specific IFN-γ-secreting CD4^+^ T cells correlate with bacterial load rather than with degree of protection in MTB-infected mice [@pone.0011019-Mittrcker1]. The massive deleterious inflammation observed in the Koch phenomenon [@pone.0011019-Koch1], which involves Ag85B among other antigens, is a primary example of de-regulated cell-mediated response to mycobacterial antigens. Thus, discovering how to control non protective antigen-specific IFN-γ production without affecting secretion of protective IFN-γ is an important challenge in tuberculosis research. Regulatory T cells (Treg) limit the magnitude of effector responses. Despite the little role played by inactivation of natural CD4^+^CD25^+^ Treg cells in improving protective BCG immunity [@pone.0011019-Quinn1], several reports indicate that pathogen-specific Treg cells, mainly IL-10-secreting CD4^+^CD25^+^ Tr1, are generated following MTB infection and suppress IFN-γ cell responses in anergic TB patients [@pone.0011019-Boussiotis1]--[@pone.0011019-Chen1]. A human CD8^+^LAG3^+^CD25^+^Treg subset, which can inhibit CD4^+^ T cell responses has also been described in purified protein derivative (PPD)^+^ subjects, but the clinical significance of this lymphocyte subset is still unknown [@pone.0011019-Joosten1]. The induction of Treg cells upon vaccination with mycobacterial antigen and the immunological mechanisms which regulate their functions need to be further investigated. Co-stimulatory molecules are up-regulated by activated T cells and participate in regulation of T cell functions. 4-1BB, the murine homologue of human CD137, is a member of the TNF-receptor super-family [@pone.0011019-Kwon1] expressed primarily on antigen-receptor activated T cells [@pone.0011019-Pollok1], [@pone.0011019-Watts1]. Although 4-1BB co-stimulates primary and secondary responses of both CD8^+^ and CD4^+^ T effector cells [@pone.0011019-Cannons1]--[@pone.0011019-Wilcox1], signalling via this molecule also results in the activation of CD4^+^ Treg cells in vitro [@pone.0011019-Elpek1] and in the generation of CD8^+^ Treg cells that suppress CD4^+^ T cell function and antibody responses in vivo [@pone.0011019-Mittler1], [@pone.0011019-Seo1]. Due to their ability to activate both effector and regulatory T cells, 4-1BB/CD137 ligands are being evaluated for treatment of cancer and auto-immune diseases [@pone.0011019-Melero2]. We have previously reported that protection conferred by immunization of mice with an Ag85B-encoding plasmid DNA was lost when DNA-primed mice were boosted with adjuvant-free Ag85B protein [@pone.0011019-Palma1]. The lack of protection was associated with the expansion of an Ag85B-specific CD4^+^ T cell subset secreting elevated IFN-γ amounts that caused loss or dilution of protective low IFN-γ secreting-CD4^+^ T cells elicited by DNA immunization [@pone.0011019-Palma1]. Co-administration of Ag85B protein with the adjuvant LTK63 reduced the generation/polarization of Ag85B-specific IFN-γ secreting cells and resulted in some recovery of protection [@pone.0011019-Palma2]. Thus, manipulating the IFN-γ production by antigen-specific CD4^+^ T cells may achieve a proper balance between protection and pathology and have vaccine-relevant implication. In the present study, we asked whether it was possible to modulate antigen-specific, IFN-γ producing CD4^+^ T cells not in the induction phase, as mentioned above, but after they have been generated. Namely, we used the previously established TB model of protective Ag85B DNA- and non-protective Ag85B DNA/protein- immunizations to investigate the role played by CD8^+^ T cells and 4-1BB ligation on controlling memory Ag85B-specific IFN-γ-secreting CD4^+^ T cells. We found that both protective and non-protective Ag85B-immunization generated antigen-specific CD8^+^ T cells that suppressed Ag85B-specific IFN-γ-secreting CD4^+^ T cells. 4-1BB ligation selectively reduced the high but non-protective CD4^+^ T cell-mediated IFN-γ response in DNA-primed and protein-boosted mice without affecting the low IFN-γ response associated with protection in mice vaccinated with DNA. The 4-1BB-mediated IFN-γ inhibition selectively in spleen cells of DNA-primed and protein-boosted mice was linked to the induction of 4-1BB exclusively on CD8^+^ T cells of these mice upon Ag85B stimulation. Results {#s2} ======= CD8^+^ T cells reduce Ag85B-mediated IFN-γ production by antigen-specific CD4^+^ T cells elicited through protective and non-protective Ag85B-immunizations {#s2a} ----------------------------------------------------------------------------------------------------------------------------------------------------------- Previously, we reported that mice immunized with Ag85B-encoding DNA were protected against MTB challenge (P-mice) while adjuvant-free Ag85B protein boosting in mice primed with Ag85B-encoding DNA (NP-mice) did not confer protection (10, 36). The lack of protection in NP-mice was associated with an excessive Ag85B-induced IFN-γ production by spleen cells (10). In spleen cells of both NP- and P-mice CD4^+^ T cell depletion almost abolished the IFN-γ release indicating that IFN-γ production in response to Ag85B stimulation was mainly attributable to CD4^+^ T cells. On the other hand, depletion of CD8^+^ T cells induced a significant increase of IFN-γ secretion, suggesting that CD8^+^ T cells may exert a negative control on antigen-specific IFN-γ secreting CD4^+^ T cells (10). To verify this hypothesis, CD4^+^ and CD8^+^ T cells of NP- and P-mice were purified through negative selection and cultured on a feeder of CD3^+^ T cell-depleted spleen cells from naïve mice. CD4^+^ T cells of NP- and P-mice, (but not naïve CD4^+^ T cells used as control, data not shown) secreted IFN-γ in response to Ag85B protein stimulation maintaining the differences in magnitude observed on unfractionated spleen cells ([Fig. 1 A and B](#pone-0011019-g001){ref-type="fig"}). When purified CD8^+^ T cells of NP- or P-mice were added to their respective CD4^+^ T cells a significant inhibition of Ag85B-mediated IFN-γ secretion was observed ([Fig. 1 A and B](#pone-0011019-g001){ref-type="fig"}), indicating that CD8^+^ T cells exert suppressing activity on antigen-specific CD4^+^ T cells. Since CD8^+^ T cells recovered from naïve mice were unable to inhibit Ag85B-mediated IFN-γ secretion by CD4^+^ T cells of NP- and P- mice ([Fig. 1 A and B](#pone-0011019-g001){ref-type="fig"}), it is likely that these CD8^+^ T cells with suppressive activity were generated during Ag85B immunization. Purified CD8^+^ T cells of both naïve and Ag85B-immunized mice or cells used as feeder did not release detectable amounts of IFN-γ (data not shown). {ref-type="sec"} from spleen cells of NP-mice (panel A) or P-mice (panel B) were cultured, alone or together (using the CD4/CD8 ratio of 1∶0.6 as found in fresh spleen cells) on a feeder of CD3^+^ T cell-depleted spleen cells of naive mice (1.5×10^5^ cells/well). Untouched purified CD8^+^ T cells recovered from naïve unvaccinated (3×10^4^ cells/well) were also added to CD4^+^ T cell co-culture. Cells were re-stimulated with Ag85B protein (5 microg/ml). Culture supernatants were harvested after 3 days for IFN-g detection by a specific quantitative sandwich ELISA Kit. Data are presented as mean of 4 independent experiments. Error bars indicate SEM. The level of statistical significance for differences were determined by a two-tailed Student *t* test (\*\*, p\<0.01) between Ag85B-induced responses by CD4^+^ T cells alone and co-cultured with CD8^+^ T cells. Purified CD8^+^ T cells of both naïve and Ag85B-immunized mice or cells used as feeder did not release detectable amounts of IFN-g (data not shown).](pone.0011019.g001){#pone-0011019-g001} 4-1BB ligation inhibits Ag85B-mediated IFN-γ production by CD4^+^ T cells exclusively in spleen cells of NP-mice: CD8^+^ T cells are required for inhibition {#s2b} ------------------------------------------------------------------------------------------------------------------------------------------------------------ Since signals through the T cell co-stimulatory molecule 4-1BB expressed by activated T cells can modulate both effector and regulatory T cell functions [@pone.0011019-Cannons1]--[@pone.0011019-Seo1], the effects induced by 4-1BB ligation on Ag85B-mediated-IFN-γ production were studied in antigenic recall experiments with spleen cells of P- and NP-mice. In spleen cells of NP-mice stimulated with Ag85B protein, treatment with an agonistic anti-4-1BB mAb reduced the excessive/non-protective accumulation of IFN-γ in culture supernatants ([Fig. 2 A and B](#pone-0011019-g002){ref-type="fig"}). The lack of IFN-γ production when CD4^+^ T cell were removed from spleen cells ([Fig. 2A](#pone-0011019-g002){ref-type="fig"}) suggested that the 4-1BB-mediated inhibition affected the IFN-γ secretion by antigen-specific CD4^+^ T cells. This hypothesis was clearly confirmed by the consistent reduction in the percentage and fluorescence intensity of IFN-γ-producing CD4^+^ T cells in Ag85B-stimulated spleen cells treated with an agonistic anti-4-1BB mAb after 4 days of culture ([Fig. 2 C](#pone-0011019-g002){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). The reduction of antigen-mediated IFN-γ accumulation in supernatants peaked at 5 days (about 48% of inhibition) ([Fig. 2 B](#pone-0011019-g002){ref-type="fig"}). The inhibitory effects mediated by 4-1BB ligation required CD8^+^ T cells since it was lost when those cells were removed from the cultures ([Fig. 2 A and B](#pone-0011019-g002){ref-type="fig"}), suggesting an involvement of the suppressor CD8^+^ T cells generated during Ag85B-vaccination in this regulatory mechanisms. {ref-type="sec"} of NP-mice (panels A--D) or P-mice (panels E--H) were re-stimulated with Ag85B protein (5 µg/ml) (panels A--C, E--G) or anti-CD3ε mAb (0.1 µg/ml) (panels D and H) in the presence of agonistic anti-4-1BB mAb or rat IgG2a control Ab (5 µg/ml each). In panels A, E, D and H, the amount of IFN-γ released in the culture supernatants after 3 days was detected by a commercial ELISA kit and the data has been reported as the mean of 5 independent experiment. In panels B and F the percentage of inhibition of Ag85B-induced IFN-γ release by an agonistic anti-4-1BB mAb in unfractionated or CD8^+^ T cell-depleted spleen population after 3 or 5 days of culture has been reported as the mean of 12 independent experiments for NP-mice and of 8 independent experiments for P-mice. Rat IgG2a control Ab did not modified the cell proliferation or IFN-γ secretion after stimulation with Ag85B and anti-CD3ε mAb (data not shown). Error bars indicate SEM. The level of statistical significance for differences between groups were determined by a two-tailed Student *t* test (\*, p\<0.05; \*\*, p\<0.01). In panels C and G, spleen cells of both NP- or P-mice after 4 days of culture and an overnight incubation with brefeldin A were stained with CD4-PE-Cy5 and then intracellular stained with FITC anti-IFN-γ Ab as reported in [Material and Methods](#s4){ref-type="sec"}. Cells were then analyzed by flow cytometry. Dot plot were generated after gating on live CD4^+^ T lymphocytes and show frequency of IFN-γ-producing cells. The rat IgG2a control Ab did not significantly change Ag85B-induced IFN-γ secretion by spleen cells of NP-mice (data not shown).](pone.0011019.g002){#pone-0011019-g002} 10.1371/journal.pone.0011019.t001 ###### Effects of 4-1BB ligation in spleen cells of NP- and P-mice stimulated with Ag85B protein on IFN-γ-producing CD4^+^ T cells and on proliferation of CD4^+^ and CD8^+^ T cells. {#pone-0011019-t001-1} CD4^+^IFN-γ^+^ CD4^+^IL-10^+^ CD4^+^CFSE^low^ CD8^+^CFSE^low^ ---------------------- ---------------- ---------------- ----------------- ----------------- **NP-mice** unstimulated 0.32±0.02 0.16±0.3 4.0±0.5 3.5±0.5 Ag85B/control IgG 5.2±0.6\# 0.95±0.02\# 39.8±3\# 12.0±1\# Ag85B/anti-4-1BB mAb 2.6±0.3§ 0.67±0.03§ 24.5±0.5§ 19.3±0.9§ **P-mice** unstimulated 0.14±0.02 nd 2.7±0.4 5.2±0.9 Ag85B/control IgG 0.62±0.06\# nd 12.3±1.9\# 14.2±1.6\# Ag85B/anti-4-1BB mAb 0.4±0.13 nd 10.7±1.8 15.0±2.3 To analyze IFN-γ- or IL-10-producing CD4^+^ T cells, unfractionated spleen cells of P- or NP-mice were cultured with Ag85B protein (5 µg/ml) in the presence of IgG2a control Ab or agonist anti-4-1BB mAb (5 µg/ml). After 2 days (for IL-10 analysis) or 4 days (for IFN-γ analysis) of culture and an overnight incubation with brefeldin A cells were stained with CD4-PE-Cy5 and then intracellular stained with PE anti-IL-10 or FITC anti-IFN-γ Ab as reported in [Material and Methods](#s4){ref-type="sec"}. Cells were then analyzed by flow cytometry and the data show frequency of IFN-γ- or IL-10-producing CD4^+^ T cells. To study cell proliferation, unfractionated spleen cells of NP- or P-mice, were labelled with CFSE as described in [Materials and Methods](#s4){ref-type="sec"} and then stimulated with Ag85B protein (5 µg/ml) in the presence of agonistic anti-4-1BB mAb (5 µg/ml) or rat IgG2a control Ab (5 µg/ml) for 4 days. At this time point cells were labelled with CD4-PE and CD8-PerCP and analyzed by flow cytometry as described in [Materials and Methods](#s4){ref-type="sec"}. The percentage of CD4^+^CFSE^low^ and CD8^+^CFSE^low^ indicates the number of replicating CD4^+^ or CD8^+^ T cells, respectively. The results are presented as mean ± SEM of the percentage of CD4^+^ T cells or CD8^+^ T cells staining for the indicated markers. Pooled data of 3 independent experiments are shown. \#, § statistical significance for differences between groups determined by ANOVA and Bonferroni-type multiple t-test (\# Ag85B/control IgG vs unstimulated; § Ag85B/control IgG vs Ag85B/anti-4-1BB mAb). nd = not done. In spleen cells of P-mice, 4-1BB ligation did not reduce either the modest IFN-γ release by CD4^+^ T cells of P-mice stimulated with Ag85B protein - as indicated by ELISA measurement in culture supernatant of spleen cells ([Fig. 2 E and F](#pone-0011019-g002){ref-type="fig"}) - or the number of Ag85B-stimulated IFN-γ producing CD4^+^ T cells at 4 days of culture, as indicated by intracellular staining ([Fig. 2 G](#pone-0011019-g002){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). Therefore, although suppressor CD8^+^ T cells were generated in spleens of both NP- and P-mice, 4-1BB ligation, through CD8^+^ T cells, reduced exclusively the massive, non-protective Ag85B-specific IFN-γ-production by CD4^+^ T cells of NP-mice while did not modify the modest antigenic secretion by CD4^+^ T cells of P-mice. The capacity of 4-1BB signalling to inhibit Ag85B-mediated IFN-γ production by CD4^+^ T cells of NP-mice was antigen-restricted. Indeed, in spleen cells of both P and NP-mice, agonistic anti-4-1BB mAb treatment increased the amount of IFN-γ secreted upon activation through CD3ε engagement on T cells ([Fig. 2 D and H](#pone-0011019-g002){ref-type="fig"}). The up-regulation of CD3ε-mediated IFN-γ production affected both CD8^+^ (the more responsive population) and CD4^+^ T cells, as suggested by *in vitro* depletion studies ([Fig. 2 D and H](#pone-0011019-g002){ref-type="fig"}). Therefore, 4-1BB acted as the expected co-stimulatory molecule in view of its ability to enhance IFN-γ production by polyclonally-activated CD4^+^ and CD8^+^ T cells. Ag85B stimulation induces expression of 4-1BB on CD8^+^ T cells exclusively in spleen cells of NP-mice {#s2c} ------------------------------------------------------------------------------------------------------ In order to explain the responsiveness of spleen cells of NP-mice to 4-1BB ligation, we analyzed whether Ag85B stimulation could induce the expression of this TNF-receptor on T cells. In spleen cells of NP-mice induction of 4-1BB expression was found on a sub-population of CD4^+^ T cells when splenocytes were stimulated for 3 days with Ag85B protein ([Fig. 3 A](#pone-0011019-g003){ref-type="fig"} and [Table 2](#pone-0011019-t002){ref-type="table"}). 4-1BB expression on CD4^+^ T cells was maintained after 6 days of culture. In spleen cells of P-mice activated with Ag85B protein a smaller percentage of CD4^+^ T cells, as compared to Ag85B-activated CD4^+^ T cells of NP-mice, expressed 4-1BB after 3 days of culture. However, after 6 days of culture no significant expression of this receptor was found on CD4^+^ T cells of P-mice ([Fig. 3 A](#pone-0011019-g003){ref-type="fig"} and [Table 2](#pone-0011019-t002){ref-type="table"}). {ref-type="sec"}. Within each panel, the percentage indicates the number of cells expressing the 4-1BB receptor. Plots, gated on CD4^+^ or CD8^+^ T cells in the lymphocyte-gate, are representative of one experiments out of 3.](pone.0011019.g003){#pone-0011019-g003} 10.1371/journal.pone.0011019.t002 ###### Effects of 4-1BB ligation in spleen cells of NP- and P-mice stimulated with Ag85B protein on expression of 4-1BB receptor on CD4^+^ and CD8^+^ T cells. {#pone-0011019-t002-2} 3 days 6 days -------------- -------------- ------------- -------------- -------------- **NP-mice** unstimulated 1.5±0.5 1.1±0.05 2.3±0.3 1.7±0.4 Ag85B 11.7±0.3\*\* 3.5±0.5\*\* 13.4±1.3\*\* 11.5±1.7\*\* **P-mice** unstimulated 1.05±0.05 0.2±0.1 1.3±0.13 0.35±0.05 Ag85B 4.5±0.6\*\* 0.2±0.1 3.1±0.7 0.45±0.05 Unfractionated spleen cells of P- or NP-mice were cultured with Ag85B protein (5 µg/ml). After 3 or 6 days of culture cells were labeled with the following labelled mAb: CD4 PE-Cy7, CD8 PE-Cy7, CD3 FITC and 4-1BB PE. Cell surface molecule expression was analyzed by flow cytometry as described in [Materials and Methods](#s4){ref-type="sec"}. The results are presented as mean ± SEM of the percentage of CD4^+^ T cells or CD8^+^ T cells positive for 4-1BB. Pooled data of 3 independent experiments are shown. \*\*, p\<0.01, statistical significance for differences between groups Ag85B vs unstimulated determined by a two-tailed Student\'s t-test. In spleen cells of NP-mice Ag85B protein stimulation caused the induction of 4-1BB on a small subset of CD8^+^ T cells after 3 days of culture and the percentage of 4-1BB^+^CD8^+^ T cells even increased after 6 days of culture ([Fig. 3 B](#pone-0011019-g003){ref-type="fig"} and [Table 2](#pone-0011019-t002){ref-type="table"}). Therefore, in response to antigen, some CD4^+^ and CD8^+^ T cells of NP-mice became responsive to ligands for 4-1BB. On the other hand, no significant induction of 4-1BB was found on CD8^+^ T cells in spleen cells of P-mice activated with Ag85B protein after 3 or 6 days of culture ([Fig. 3 B](#pone-0011019-g003){ref-type="fig"} and [Table 2](#pone-0011019-t002){ref-type="table"}). Effect of 4-1BB ligation on modulation of CD4^+^ and CD8^+^ T cell surface marker expression in Ag85B-activated spleen cells of NP mice {#s2d} --------------------------------------------------------------------------------------------------------------------------------------- Next, the effects of 4-1BB ligation on the phenotype of both CD4^+^ and CD8^+^ T cells was studied in culture of spleen cells of NP- or P- mice stimulated with Ag85B protein. In particular, the expression of T cell activation markers expressed upon TCR complex stimulation such as CD25, the IL-2Rα, CD38, an ectoenzyme with cyclase and hydrolase enzymatic activity, CD233, an MHC class II binding CD4 homologue, and CD44, an adhesion molecule expressed by activated memory T cells, was analyzed. All these markers are up-regulated by both activated effector and regulatory cells [@pone.0011019-Li1], [@pone.0011019-Chen1], [@pone.0011019-Elpek1]--[@pone.0011019-Seo1], [@pone.0011019-Lund1]--[@pone.0011019-Read1]. In spleen cells of NP-mice Ag85B protein stimulation induced up-regulation of CD25, CD38, CD233 and CD44 on about 40% of CD4^+^ T cells ([Table 3](#pone-0011019-t003){ref-type="table"}). Likely these CD4^+^ T cells were effectory memory cells in view of the low expression of CD62L (data not shown). Upon agonistic anti-4-1BB mAb treatment, CD4^+^ T cells of NP-mice down-modulated surface expression of all activated T cell markers, suggesting a reduced activation of antigen-specific CD4^+^ T cells following agonistic anti-4-1BB mAb treatment. 10.1371/journal.pone.0011019.t003 ###### Effects of 4-1BB ligation in spleen cells of NP- and P-mice stimulated with Ag85B protein on expression of activation markers on CD4^+^ T cells. {#pone-0011019-t003-3} CD4^+^ gated T lymphocytes ---------------------- ------------ ---------------- ------------- ------------ **NP-mice** unstimulated 14.0±2.0 17.0±0.6 10.0±1.0 15.5±0.5 Ag85B/control IgG 40.5±0.5\# 44.6±1.3\# 49±3.0\# 41.5±0.5\# Ag85B/anti-4-1BB mAb 30.5±1.5§ 29.6±0.9§ 35.5±1.5§ 26.5±0.5§ **P-mice** unstimulated 10.6±0.6 16.0±0.1 5.5±0.5 14.0±0.1 Ag85B/control IgG 20.7±0.6\# 24.6±0.3\# 14.5±0.5\# 21.0±1.0\# Ag85B/anti-4-1BB mAb 23.0±1.0 25.0±1.7 14.0±1.0 22.5±0.5 Unfractionated spleen cells of P- or NP-mice were cultured with Ag85B protein (5 µg/ml) in the presence of IgG2a control Ab or agonist anti-4-1BB mAb (5 µg/ml). After 6 days of culture cells were labeled with the indicated mAbs. Cell surface molecule expression was analyzed by flow cytometry as described in [Materials and Methods](#s4){ref-type="sec"}. The results are presented as mean ± SEM of the percentage of CD4^+^ T cells staining for the indicated markers. Pooled data of 3 independent experiments are shown. \#, § statistical significance (p\<0.05 or \<0.01) for differences between groups determined by ANOVA and Bonferroni-type multiple t-test (\# Ag85B/control IgG vs unstimulated; § Ag85B/control IgG vs Ag85B/anti-4-1BB mAb). In spleen cells of P-mice, Ag85B stimulation induced a weak up-regulation of CD25, CD38, CD233 and CD44 expression on CD4^+^ T cells ([Table 3](#pone-0011019-t003){ref-type="table"}). These data were consistent with a reduced activation of CD4^+^ T cells of P-mice compared to CD4^+^ T cells of NP-mice upon Ag85B protein stimulation. Moreover, agonistic anti-4-1BB mAb treatment did not influence the phenotype of Ag85B-activated CD4^+^ T cells. In spleen cells of both P- and NP-mice Ag85B protein stimulation led to up-regulation of CD233 on a subset of CD8^+^ T cells, while no changes were observed in the percentage of cells expressing CD25, CD38 and CD44 ([Table 4](#pone-0011019-t004){ref-type="table"}). Agonistic anti-4-1BB mAb treatment induced down-modulation of CD233 exclusively on CD8^+^ T cells of NP-mice. 10.1371/journal.pone.0011019.t004 ###### Effects of 4-1BB ligation in spleen cells of NP- and P-mice stimulated with Ag85B protein on expression of activation markers on CD8^+^ T cells. {#pone-0011019-t004-4} CD8^+^ gated T lymphocytes ---------------------- ---------- ---------------- ------------- ---------- **NP-mice** unstimulated 10.5±1.5 12.4±1.0 14.4±1.2 7.3±1.4 Ag85B/control IgG 9.6±2.1 13.3±1.0 32.8±1.1\# 8.6±1.6 Ag85B/anti-4-1BB mAb 9±1.2 13.1±1.0 22.4±1.3§ 12.5±1.2 **P-mice** unstimulated 8.0±1.5 15.5±1.6 6.0±1.0 9.7±0.8 Ag85B/control IgG 6.6±1.2 15.5±0.2 15.3±0.3\# 7.7±0.9 Ag85B/anti-4-1BB mAb 6.7±0.8 16±1.0 15.5±0.5 8.9±1.0 Unfractionated spleen cells of P- or NP-mice were cultured with Ag85B protein (5 µg/ml) in the presence of IgG2a control Ab or agonist anti-4-1BB mAb (5 µg/ml). After 6 days of culture cells were labeled with the indicated mAbs. Cell surface molecule expression was analyzed by flow cytometry as described in [Materials and Methods](#s4){ref-type="sec"}. The results are presented as mean ± SEM of the percentage of CD8^+^ T cells staining for the indicated markers. Pooled data of 3 independent experiments are shown. \#, § statistical significance (p\<0.05 or \<0.01) for differences between groups determined by ANOVA and Bonferroni-type multiple t-test (\# Ag85B/control IgG vs unstimulated; § Ag85B/control IgG vs Ag85B/anti-4-1BB mAb). 4-1BB ligation has opposite effects on cell proliferation in spleen of NP-mice stimulated with Ag85B: reducing proliferation of CD4^+^ T cells while enhancing that of CD8^+^ T cells {#s2e} ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- In order to analyze the effects of 4-1BB ligation on CD4^+^ and CD8^+^ T cell proliferation of NP- and P-mice, the CFSE labelled spleen cells were stimulated with Ag85B protein for 4 days. In spleen cells of NP-mice, Ag85B protein stimulation induced a remarkable CD4^+^ T cell proliferation as indicated by CFSE dilution ([Fig. 4](#pone-0011019-g004){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). Treatment with an agonistic anti-4-1BB mAb significantly reduced the number of proliferating CD4^+^ T cells ([Fig. 4](#pone-0011019-g004){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). These data are consistent with the reduced expression of activation markers observed on Ag85B-activated CD4^+^ T cells of NP-mice following 4-1BB ligation ([Table 3](#pone-0011019-t003){ref-type="table"}). Ag85B protein stimulation also induced CD8^+^ T cell proliferation. However, the number of proliferating CD8^+^ T cells increased in the presence of the agonistic anti-4-1BB mAb treatment, suggesting that the subset of CD8^+^ T cells expressing 4-1BB may be expanded by a direct engagement of this receptor ([Fig. 4](#pone-0011019-g004){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). {ref-type="sec"} and then stimulated with Ag85B protein (5 µg/ml) in the presence of agonistic anti-4-1BB mAb (5 µg/ml) or rat IgG2a control Ab (5 µg/ml) for 4 days. At this time point cells were labelled with CD4-PE and CD8-PerCP and analyzed by flow cytometry as described in [Materials and Methods](#s4){ref-type="sec"}. The percentage indicates the number of replicating CD4^+^ or CD8^+^ T cells. Plots, gated on CD4^+^ or CD8^+^ T cell in the lymphocyte-gate, are representative of one experiments out of 3.](pone.0011019.g004){#pone-0011019-g004} In spleen cells of P-mice, Ag85B-stimulation induced a lower CD4^+^ T cell proliferation as compared to that observed in CD4^+^ T cells of NP-mice while the number of proliferating CD8^+^ T cells were similar in the two groups of immunized mice ([Fig. 4](#pone-0011019-g004){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). Agonistic anti-4-1BB mAb treatment had only a little influence on CD4^+^ or CD8^+^ T cell proliferation in P-mice ([Fig. 4](#pone-0011019-g004){ref-type="fig"} and [Table 1](#pone-0011019-t001){ref-type="table"}). These data are consistent with the low or absent expression of 4-1BB on CD4^+^ T and CD8^+^ T cells of P-mice. Effects of 4-1BB ligation on cytokine/chemokine production in Ag85B-stimulated spleen cells of NP-mice {#s2f} ------------------------------------------------------------------------------------------------------ We next asked whether 4-1BB signalling could modulate Ag85B-mediated production of other cytokines, besides IFN-γ, playing a recognized positive or negative role in protection against MTB. Cytokine/chemokines such IL-10, TGF-β, MIP-1β and XCL-1 capable to inhibit the IFN-γ production by CD4^+^ T cells [@pone.0011019-Joosten1], [@pone.0011019-Roncarolo1]--[@pone.0011019-Ordway1] were also analyzed to address the mechanisms of 4-1BB-mediated IFN-γ inhibition. In spleen cells of NP-mice, the low Ag85B-induced secretion of IL-2 and TNF-α, Th1 associate cytokines was preserved in the presence of agonist anti-4-1BB mAb ([Fig. 5 A](#pone-0011019-g005){ref-type="fig"}). Also IL-6 release induced by Ag85B protein stimulation was unaffected by agonist anti-4-1BB mAb treatment ([Fig. 5 A](#pone-0011019-g005){ref-type="fig"}). 4-1BB ligation did not induce production of IL-4, a Th2 cytokine, IL-17, a cytokine which contribute to protection to MTB challenge [@pone.0011019-Khader1], [@pone.0011019-Werninghaus1] and XCL-1, a chemokine released by suppressor CD8^+^ T cells during the chronic stage of infection with MTB that negatively affects production of IFN-γ by CD4^+^ T cells [@pone.0011019-Ordway1]. None of the cytokines above were released upon Ag85B stimulation ([Fig. 5 A](#pone-0011019-g005){ref-type="fig"}). Although no further secreted in response to antigenic stimulation, some TGF-β, a potent cytokine suppressing Th1 function [@pone.0011019-Workman1], was nonetheless found in spleen cell cultures ([Fig. 5 A](#pone-0011019-g005){ref-type="fig"}). Hence, to exclude any involvement of TGF-β in 4-1BB-mediated inhibition of IFN-γ, an anti-TGFβ neutralizing mAb was added to unfractionated spleen cell culture of NP-mice stimulated with Ag85B protein and agonistic anti-4-1BB mAb. Also in this condition, the ability of 4-1BB ligation to reduce Ag85B-induced IFN-γ secretion was unchanged ([Fig. 5 E](#pone-0011019-g005){ref-type="fig"}). {ref-type="sec"} (panels B and C), were re-stimulated with Ag85B protein (5 µg/ml) in the presence of agonistic anti-4-1BB mAb (5 µg/ml) or rat IgG2a control Ab (5 µg/ml). Culture supernatants were harvested at 48 h for IL-2, IL-4, at 72 h for IFN-γ, IL-6, IL-10, MIP-1β, TGF-β1, TNF-α, XCL-1 and at 96 h for IL-17 detection by specific quantitative sandwich ELISA Kits. Data are presented as mean of 3 independent experiments for panels A, B, and C. Error bars indicate SEM. The level of statistical significance for differences between Ag85B stimulation in the presence of IgG2a control Ab or agonistic anti-4-1BB mAb were determined by a two-tailed Student *t* test (\*, p\<0.05; \*\*, p\<0.01). In panel D, spleen cells after 2 days of culture and an overnight incubation with brefeldin A were stained with CD4-PE-Cy5 and then intracellular stained with PE anti-IL-10 Ab as reported in [Material and Methods](#s4){ref-type="sec"}. Cells were then analyzed by flow cytometry. Dot plot were generated after gating on live CD4^+^ T lymphocytes and show frequency of IL-10-producing cells. In panel E, the IDO-inhibitor, 1-MT (400 µM), or neutralizing anti-TGF-β1.2 mAb (2 µg/ml) were added simultaneously to culture of spleen cells stimulated with Ag85B protein (5 µg/ml) in the presence of agonistic anti-4-1BB mAb (5 µg/ml) or rat IgG2a control Ab (5 µg/ml). Culture supernatants were harvested at 72 h for IFN-γ detection by specific quantitative sandwich ELISA Kits. Percent of 4-1BB-mediated inhibition of Ag85B-induced IFN-γ release is shown in round brackets.](pone.0011019.g005){#pone-0011019-g005} On the other hand, 4-1BB ligation decreased the accumulation of IL-10 and MIP-1β ([Fig. 5 B and C](#pone-0011019-g005){ref-type="fig"}). Since soluble forms of IL-10 and MIP-1β can both control IFN-γ production [@pone.0011019-Boussiotis1], [@pone.0011019-Chen1], [@pone.0011019-Joosten1], [@pone.0011019-Roncarolo1] the reduced level of these factors found following 4-1BB ligation did not fit with a their involvement in 4-1BB-mediated IFN-γ inhibition. IL-10 was exclusively produced by CD4^+^ T cells upon Ag85B stimulation in total spleen cells ([Fig. 5 C and D](#pone-0011019-g005){ref-type="fig"}) and 4-1BB-ligation reduced IL-10 production by Ag85B-stimulated CD4^+^ T cells as shown in intracellular staining ([Fig. 5 D](#pone-0011019-g005){ref-type="fig"}). IL-10-producing CD4^+^ T cells did not produce IFN-γ (data not shown), indicating that in spleen cell populations two different Ag85B-responding CD4^+^ T cell subsets secreting IL-10 or IFN-γ were present. Since secretion of MIP-1β was in part due to cells other than CD4^+^T-cells ([Fig. 5 B](#pone-0011019-g005){ref-type="fig"}), the possibility of MIP-1β secretion by CD8^+^ T cells was supposed. In any case 4-1BB ligation did not modulate the release of MIP-1β by non-CD4^+^ T cells ([Fig. 5 B](#pone-0011019-g005){ref-type="fig"}). Moreover, unlike IFN-γ, CD4^+^ T cell-mediated production of IL-10 and MIP-1β decreased in a CD8^+^ T cell-independent manner upon 4-1BB ligation ([Fig. 5 B and C](#pone-0011019-g005){ref-type="fig"}), suggesting that in Ag85B-responding CD4^+^ T cells the regulatory mechanisms controlling secretion of IFN-γ differ compare to those modulating other cytokines influential in TB protection. In addition to cytokines, some metabolic products which affect tryptophan metabolism play a role in inhibition of T cell responses. In particular, the tryptophan catabolising enzyme indoleamine-pyrrole 2,3 dioxygenase (IDO) leads to the inhibition of CD4^+^ T cell responses when up-regulated by an agonistic 4-1BB mAb in macrophages and dendritic cells [@pone.0011019-Seo1], [@pone.0011019-Choi1]. To test IDO functions we used 1-methyltryptophan (1-MT), which abrogates 4-1BB-controlled IDO-mediated immune suppression [@pone.0011019-Choi1]. When 1-MT was added to unfractionated spleen cell culture of NP-mice, the ability of 4-1BB ligation to reduce Ag85B-induced IFN-γ secretion was unchanged ([Fig. 5 E](#pone-0011019-g005){ref-type="fig"}). However, 1-MT treatment induced increased IFN-γ levels in Ag85B-stimulated cells independent of treatment with an agonistic anti-4-1BB mAb ([Fig. 5 E](#pone-0011019-g005){ref-type="fig"}). This was presumably caused by the high level of IFN-γ in culture, since IFN-γ up-regulates IDO in macrophages and dendritic cells (34). In spleen cells of P-mice, treatment with the agonistic anti-4-1BB mAb did not significantly modify Ag85B-induced release of IL-2, IL-6, MIP-1β and TNF-α, similarly to IFN-γ production, and did not induce release of IL-10, IL-4, TGF-β, IL-17 and XCL-1 ([Table 5](#pone-0011019-t005){ref-type="table"}), supporting again the unresponsiveness of these T cells to ligands for 4-1BB. 10.1371/journal.pone.0011019.t005 ###### Effects of 4-1BB ligation in Ag85B-activated spleen cells of P-mice on cytokine/chemokine production. {#pone-0011019-t005-5} medium control Ag85B/control IgG Ag85B/anti-4-1BB mAb --------- ---------------- ------------------- ---------------------- IL-2 0±0 144±60 97±50 IL-4 0±0 0±0 0±0 IL-6 0±0 41±14 37±17 IL-10 0±0 0±0 0±0 IL-17 0±0 0±0 0±0 TNF-α 0±0 32±6 44±6 MIP-1 β 30±20 2436±195 2770±140 TGF−β1 713±102 702±98 768±130 XCL-1 0±0 0±0 0±0 Unfractionated spleen cells of P-mice were cultured with Ag85B protein (5 µg/ml) in the presence of IgG2a control Ab or agonist anti-4-1BB mAb (5 µg/ml). Culture supernatants were harvested at 48 h for IL-2, IL-4; at 72 h for IL-6, IL-10, XCL-1, MIP-1β, TGF-β1, TNF-α and at 96 h for IL-17 detection by specific quantitative sandwich ELISA Kits. The results are presented as mean ± SEM in pg/ml for cytokine detection. Pooled data of 3 independent experiments are shown. 4-1BB ligation on purified CD4^+^ T cells of NP-mice stimulated with Ag85B protein inhibits release of IL-10 and MIP-1β, but not IFN-γ: requirement of CD8^+^ T cells for 4-1BB-mediated IFN-γ inhibition {#s2g} --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Since Ag85B-stimulated CD4^+^ T cells of NP-mice expressed the 4-1BB molecule, the effects induced by a direct engagement of this TNF-receptor on these cells were investigated. Therefore, antigenic recall studies were performed on purified CD4^+^ T cells cultured on a feeder of CD3^+^ T cell-depleted spleen cells of naive mice. In addition to IFN-γ (as previously shown in [Fig. 1](#pone-0011019-g001){ref-type="fig"}), IL-10 and MIP-1β were also secreted in such co-culture upon Ag85B protein stimulation ([Fig. 6 A--C](#pone-0011019-g006){ref-type="fig"}). Treatment with an agonistic anti-4-1BB mAb, while increasing IFN-γ release, significantly reduced CD4^+^ T cell-mediated-secretion of IL-10 and MIP-1β ([Fig. 6 A--C](#pone-0011019-g006){ref-type="fig"}).The effects induced by 4-1BB ligation on release of these cytokines were attributable to CD4^+^ T cells since CD3^+^ T cell-depleted spleen cells of naive mice used as feeder did not release IFN-γ and IL-10 in response to Ag85B protein and/or to agonistic anti-4-1BB mAb, and released only a low amount of MIP-1β not modified by 4-1BB treatment ([Fig. 6 A--C](#pone-0011019-g006){ref-type="fig"}). As previously shown in [Fig. 1A](#pone-0011019-g001){ref-type="fig"}, addition of purified CD8^+^ T cells of NP-mice to the co-culture significantly reduced IFN-γ secretion by Ag85B-stimulated CD4^+^ T cells ([Fig. 6 A](#pone-0011019-g006){ref-type="fig"}). Treatment of these CD4^+^/CD8^+^ T cell-co-culture with the agonistic anti-4-1BB mAb resulted in a further IFN-γ inhibition (50.2±3.5 versus 36±3 in the presence or absence of 4-1BB ligation, respectively) ([Fig. 6 A](#pone-0011019-g006){ref-type="fig"}), suggesting that 4-1BB ligation on 4-1BB^+^CD8^+^ T cells enhanced the suppressive effects exerted by CD8^+^ T cells on CD4^+^ T cell-mediated IFN-γ release. A low, almost undetectable release of IFN-γ by purified CD8^+^ T cells was found upon Ag85B stimulation, and IFN-γ secretion was not significantly modified by 4-1BB ligation, further confirming that IFN-γ released by spleen cells upon antigenic recall was mainly due to CD4^+^ T cells. Differently from what observed for IFN-γ release, the presence of CD8^+^ T cells neither affected release of IL-10 and MIP-1β by Ag85B-stimulated CD4^+^ T cells nor modified the reduction of IL-10 and MIP-1β secretion caused by a direct engagement of 4-1BB on Ag85B-activated CD4^+^ T cells ([Fig. 6 B and C](#pone-0011019-g006){ref-type="fig"}). Purified CD8^+^ T cells released low amount of MIP-1β upon Ag85B stimulation that remained unaltered following 4-1BB ligation while IL-10 was undetectable in CD8^+^ T cell culture supernatants ([Fig. 6 B and C](#pone-0011019-g006){ref-type="fig"}). {ref-type="sec"} from spleen cells of NP-mice were cultured, alone or together (using the CD4/CD8 ratio of 1∶0.6 as found in fresh spleen cells) on a feeder of CD3^+^ T cell-depleted spleen cells of naive mice (1.5×10^5^ cells/well). Cells were re-stimulated with Ag85B protein (5 µg/ml) in the presence of agonistic anti-4-1BB mAb or rat IgG2a control Ab (5 µg/ml each). Culture supernatants were harvested at at 72 h for IFN-γ, (panel A), MIP-1β (panel B) and IL-10 (panel C) detection by specific quantitative sandwich ELISA Kits. Data are presented as mean of 4 independent experiments. Error bars indicate SEM. The level of statistical significance for differences were determined by a two-tailed Student\'s *t*-test (\*, p\<0.05; \*\*, p\<0.01) between Ag85B stimulation in the presence of rat IgG2a control Ab or agonistic anti-4-1BB mAb (panels A--C) and between Ag85B-induced responses (both in the presence or absence of agonist anti-4-1BB mAb) by CD4^+^ T cells alone and co-cultured with CD8^+^ T cells ( panel A).](pone.0011019.g006){#pone-0011019-g006} Antigenic recall experiments performed with purified CD4^+^ and/or CD8^+^ T cells of P-mice indicated that treatment with agonistic anti-4-1BB mAb did not influence secretion of IFN-γ and MIP-1β by CD4^+^ T cells nor the CD8^+^T cell-mediated inhibition of IFN-γ release by CD4^+^ T cells (data not shown). These results are in agreement with the lack o low 4-1BB expression on Ag85B-stimulated T cells of P-mice. In vivo antigenic stimulation induces expression of 4-1BB on Ag85B-specific CD8^+^ T cells and 4-1BB ligation inhibits the antigen-mediated IFN-γ production and expansion of CD4^+^ T cells from NP-mice {#s2h} --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- To evaluate the effects of *in vivo* 4-1BB triggering on Ag85B-specific CD4^+^ and CD8^+^ T cells, adoptive transfer experiments with spleen cells of NP-mice were performed. Spleen cells from NP-mice (CD45.2^+^) were injected into C57BL/6Ly5.1 recipient mice (expressing CD45.1 leukocyte antigen). On post-injection (p.i.) day 1, mice were inoculated s.c. with 10 µg of recombinant Ag85B protein and then i.p. with 150 µg agonistic anti-mouse 4-1BB mAb or with 150 µg rat IgG2a isotype control mAb. On p.i. day 3, mice were again injected with 150 µg agonistic anti-mouse 4-1BB mAb or with 150 µg rat IgG2a isotype control Ab. As controls, a group of mice adoptively transferred with splenocytes from NP-mice was treated with neither antigen Ag85B nor mAbs, and two groups of mice, not inoculated with splenocytes from NP-mice, received only the treatment with Ag85B and mAbs as descrived above. On p.i. day 10, all mice were sacrificed and spleens were harvested for immunological studies ([Fig. 7 A](#pone-0011019-g007){ref-type="fig"}). {ref-type="sec"}, after 4 days of culture were stained with CD4-PE, CD8-PerCP and CD45.2-APC and analyzed by flow cytometry. The percentage indicates the number of replicating CD4^+^CD45.2^+^or CD8^+^CD45.2^+^ T cells. In panel E the frequency of IFN-γ-producing CD4^+^CD45.2^+^ T cells were measured by FACS analysis after 4 days of culture and an overnight incubation with brefeldin A on cells labelled with CD4-PerCP-Cy5.5, CD45.2-FITC and then intracellular stained with PE anti-IFN-γ Ab. Cells were then analyzed by flow cytometry. In panel F the amount of IFN-γ released in the culture supernatants after 4 days was detected by a commercial ELISA kit. The data has been reported as the mean of 3 individual determinations for each group. Error bars indicate SEM. The level of statistical significance for differences between unstimulated and *ex vivo* antigen-stimulated spleen cells of mice adoptively transferred with cells from NP-mice and treated *in vivo* with Ag85B protein in the presence of IgG2a control Ab or agonistic anti-4-1BB mAb were determined by a two-tailed Student *t* test (\*, p\<0.05; \*\*, p\<0.01).](pone.0011019.g007){#pone-0011019-g007} In spleen cells of mice adoptively transferred with cells from NP-mouse the frequency of CD4^+^ and CD8^+^ T cells originating from NP-mice following antigenic stimulation *in vivo* with Ag85B protein in the presence of anti-4-1BB mAb or control IgG2a mAbs was analyzed by gating lymphocytes positive for CD45.2. We found that injection of Ag85B into mice induced a significant expansion of Ag85B-specific CD4^+^ T cells ([Fig. 7 B](#pone-0011019-g007){ref-type="fig"}, [Table 6](#pone-0011019-t006){ref-type="table"}). Moreover, *in vivo* Ag85B treatment, induced the expression of 4-1BB on both CD4^+^ and CD8^+^ T cells ([Fig. 7 C](#pone-0011019-g007){ref-type="fig"}, [Table 6](#pone-0011019-t006){ref-type="table"}), suggesting that *in vivo* 4-1BB engagement could modulate their responses. In fact, treatment with anti-4-1BB mAb *in vivo* significantly reduced the percentage of CD4^+^ T cells responding to Ag85B treatment. Although the increase of CD8^+^ T cell frequency was not statistically significant, *in vivo* 4-1BB ligation significantly reversed the ratio of CD4^+^ and CD8^+^ T cells (0.64±0.3 vs 1.2±0.01 in mice injected with Ag85B and anti-4-1BB mAb compared with mice injected with Ag85B and control IgG2a mAb, respectively ([Fig. 7 B](#pone-0011019-g007){ref-type="fig"}, [Table 6](#pone-0011019-t006){ref-type="table"}). This indicates an expansion of CD8^+^ T cells at the expense of CD4^+^ T cells in mice treated with anti-4-1BB mAb. To better clarify whether *in vivo* 4-1BB ligation downgrades the action of Ag85B-specific CD4^+^ T cells while expanding CD8^+^ T cells, we also performed *ex vivo* antigenic recall experiments with spleen cells of mice adoptively transferred with cells from NP-mouse. After 4 days of culture the percentage of proliferating CD8^+^CD45.2^+^T cells, both in unstimulated and *ex vivo* Ag85B-restimulated splenocytes, was significantly enhanced in anti-4-1BB mAb-treated mice ([Fig. 7 D](#pone-0011019-g007){ref-type="fig"}). On the contrary, the anti-4-1BB mAb *in vivo* treatment significantly reduced the frequency of proliferating CD4^+^CD45.2^+^T cells ([Fig. 7 D](#pone-0011019-g007){ref-type="fig"}) as well as the percentage of IFN-γ-producing CD4^+^CD45.2^+^T cells responding to *ex vivo* antigenic re-stimulation ([Fig. 7 E](#pone-0011019-g007){ref-type="fig"}). Moreover, IFN-γ accumulation in culture supernatants, both in unstimulated and re-stimulated ex vivo with Ag85B protein spleen cell culture, was significantly reduced in anti-4-1BB mAb-treated mice ([Fig. 7 F](#pone-0011019-g007){ref-type="fig"}). The IFN-γ released in the cultures was mainly released by cells originating from NP-mice ([Fig. 7 F](#pone-0011019-g007){ref-type="fig"}). Altogether these data indicate that *in vivo* 4-1BB ligation reduced antigen-mediated-activation of Ag85B-specific CD4^+^ T cells while expanding CD8^+^ T cells of NP-mice, in keeping with the results of the *in vitro* experiments (see above). 10.1371/journal.pone.0011019.t006 ###### Effects of Ag85B stimulation and 4-1BB ligation on in vivo frequency of CD4^+^ and CD8^+^ T cells and 4-1BB expression of cells from NP-mice adoptively transferred into C57BL/6Ly5.1 recipient mice. {#pone-0011019-t006-6} CD45.2^+^ gated T lymphocytes ---------------------- ------------ ----------------- --------------- ------------ none 21.2±0.5 21.8±0.5 2.8±0.1 2.4±0.1 Ag85B/control IgG 25.8±0.3\# 22.1±0.8 10.1±1.0\# 15.4±0.9\# Ag85B/anti-4-1BB mAb 16.7±0.5§ 25.8±1.3 7.1±0.6§ 12.3±0.8 C57BL/6Ly5.1 mice adoptively transferred with 10^7^ spleen cells from NP-mice (cells CD45.2^+^) were treated or not with antigen Ag85B and mAbs as described in [Materials and Methods](#s4){ref-type="sec"}. Slpeen cells were harvested on post-infection day 10, stained with CD45.2-FITC, CD4-PE, CD8-PerCP, CD4PerCPCy5.5 and 4-1BB-PE mAbs and then analyzed by flow cytometry. The data, obtained by gating CD45.2^+^ cells on lymphocyte population, show the frequency of CD4^+^ and CD8^+^ T cells originating from NP-mice and the percentage of T cells expressing 4-1BB. The results are presented as mean ± SEM of 3 mice per group. \#, § statistical significance for differences between groups determined by ANOVA and Bonferroni-type multiple t-test (\# Ag85B/control IgG vs none; § Ag85B/control IgG vs Ag85B/anti-4-1BB mAb). Discussion {#s3} ========== Dampening the drawbacks of elevated antigen-specific IFN-γ secretion by CD4^+^ T cells without affecting the protective IFN-γ response is a difficult challenge in TB research, particularly to generate a strongly immunogenic though safe vaccine. In a murine model of TB, combined DNA/protein immunization with the mycobacterial antigen Ag85B (NP-mice) increases IFN-γ production by antigen-specific CD4^+^ T cells but reduces protection from challenge relative to the marked protection seen in mice immunized with DNA alone (P-mice) (10, 36). Here we report that both protective and non-protective Ag85B-immunizations generate antigen-specific CD8^+^ T cells which, in *ex vivo* experiments, suppress IFN-γ-production by Ag85B-specific CD4^+^ T cells. Signals via the TNF-receptor 4-1BB down-modulate, via CD8^+^ T cells, the high-level of IFN-γ production by Ag85B-specific CD4^+^ T cell associated with poor protection in NP-mice while preserving the low IFN-γ response that is associated with protective immunity in P-mice. The observed inhibition of memory IFN-γ-secreting CD4^+^ T cell in NP-mice was likely due to the induction, upon Ag85B stimulation, of 4-1BB expression selectively on suppressor CD8^+^ T cells of NP-mice. Thus, the 4-1BB capability of discriminating between low and high antigen-specific CD4^+^ T-cell-mediated IFN-γ responses could be instrumental to find novel strategies based on T cells and co-stimulatory signals to down-regulate the effects of elevated antigen-specific IFN-γ responses associated with TB pathology and its negative impact on vaccine protection. As a matter of fact, knowledge of T-cell co-stimulatory molecules on MTB immunity and their potential use as new therapeutic targets is still lacking. In addition, *in vivo* 4-1BB ligation reduced the activation, IFN-γ production and expansion of CD4^+^ T cells of NP-mice re-stimulated *in vivo* with Ag85B, suggesting that the inhibitory phenomenon observed *in vitro* can also occur *in vivo*. CD8^+^ T cells have a recognized, direct role in protection against TB through their cytotoxic activity on MTB-infected cells [@pone.0011019-Stenger1]. The main finding of this paper strongly indicate that CD8^+^ T cells may have an additional role in TB as regulator of CD4^+^ T cell-mediated responses, which are mainly implicated in protection against MTB infection [@pone.0011019-Cooper1], [@pone.0011019-Jung1]. A role for CD8^+^ Treg cells was postulated to explain increased IFN-γ production by antigen-specific murine CD4^+^ T cells following CD8^+^T cell-depletion *ex vivo* [@pone.0011019-Palma1] and in vivo [@pone.0011019-Ordway1]. Here we provide the first evidence of antigen-induced suppressor CD8^+^ T cells generated in vivo that inhibit CD4^+^ T cell proliferation and IFN-γ secretion *ex vivo*. These suppressor CD8^+^ T cells were found in the spleen of both NP- and P-mice suggesting that they are an intrinsic component of the response to immunization. Indeed, splenic CD8^+^ T cells of naive mice did not show any regulatory activity. Moreover, while suppressor CD8^+^ T cells were generated both in P- and NP-mice, 4-1BB ligation suppressed the CD4^+^ T cell-mediated IFN-γ response only in NP-mice. This suppressive effect was linked to the induction, upon Ag85B protein stimulation, of 4-1BB receptor expression only on CD8^+^ T cells of NP-mice. Thus, CD8^+^ T cells of NP-mice responded to 4-1BB ligands while CD8^+^ T cells of P-mice were insensitive to 4-1BB ligation because they did not express 4-1BB molecules. Consistent with these results is the enhancement of antigen-specific cell proliferation of CD8^+^ T cells of NP-mice but not of P-mice following agonistic anti-4-1BB mAb treatment. The elevated level of IFN-γ found in spleen cell cultures of NP-mice upon Ag85B protein stimulation may be responsible for the selective expression of 4-1BB on the surface of CD8^+^ T cells of NP-mice. In fact, during chronic inflammation, 4-1BB receptor expression was up-regulated on CD8^+^ T cell surface by inflammatory mediators including IFN-γ [@pone.0011019-Olofsson1], [@pone.0011019-Giscombe1]. Therefore expression of 4-1BB on suppressor CD8^+^ T cells in mice producing elevated amount of IFN-γ may be a self-regulatory mechanism that curtails excessive, antigen-driven IFN-γ response in MTB infection. This hypothesis is in agreement with the high expression of 4-1BB induced by Ag85B stimulation *in vivo* on CD8^+^ T cells of NP-mice and expansion of these cells upon *in vivo* anti-4-1BB mAb treatment. The role played by 4-1BB in regulating CD8^+^ Treg cells deserves to be further investigated in TB and also in other infectious diseases. In fact, induction of pathogen-specific CD8^+^ Treg cells, involved in regulating both pathogen eradication and immunopathology [@pone.0011019-Mills1], [@pone.0011019-Belkaid1], have been described in various bacterial [@pone.0011019-Jones1], viral [@pone.0011019-Accapezzato1]--[@pone.0011019-Haynes1] and parasitic infections [@pone.0011019-Ormer1]. The molecular mechanisms underlying 4-1BB mediated restriction of IFN-γ production remain to be elucidated. Apparently, none of the mechanisms thus far proposed to explain inhibition of IFN-γ secreting CD4^+^ T cells applies to our findings. First, soluble MIP-1β secreted by CD8^+^CD233^+^CD25^+^ Treg has been proposed to suppress antigen-mediated CD4^+^ T cell responses in PPD^+^ responders via inhibition of Ca^2+^ flux [@pone.0011019-Joosten1]. Even though we found release of MIP-1β by CD8^+^ T cells upon Ag85B stimulation and 4-1BB ligation, the inhibitory action of soluble MIP-1β did not match MIP-1β release by Ag85B-activated CD4^+^ T cells nor the reduced, total MIP-1β accumulation in spleen cells upon 4-1BB ligation. In addition, the high CD38 expression on CD4^+^ T cells might have counteracted MIP-1β inhibition of Ca^2+^ flux by activating Ca^2+^ influx through an ADPR-gated cation channel [@pone.0011019-Lund1]. Second, release of IL-10 and TGF-β, which are main mediators of CD8^+^ and CD4^+^ Treg cells [@pone.0011019-Roncarolo1], [@pone.0011019-Zhang1], [@pone.0011019-Mills1] are not relevant for suppression of IFN-γ production in our experiments. Agonist anti-4-1BB mAb did not increase the basal TGF-β accumulation found in spleen cultures and a neutralizing anti-TGF-β mAb was ineffective in contrasting 4-1BB effects on IFN-γ release. 4-1BB engagement did not induce secretion of IL-10 by CD8^+^ T cells, and even down-regulated the secretion of IL-10 by CD4^+^ T cells. Taking into account that IL-10-secreting CD4^+^ Treg isolated from TB patients have been often considered a poor prognostic factor [@pone.0011019-Boussiotis1]--[@pone.0011019-Chen1], the IL-10-independent mechanisms of IFN-γ regulation described in this paper may constitute a protective mechanism that limits dangerous inflammation without being associated with the anergy induced by IL-10 in TB patients. Third, XCL-1, a chemokine produced by activated CD8^+^ T cells during the chronic stage of MTB infection, which negatively affects production of IFN-γ by CD4^+^ T cells [@pone.0011019-Ordway1], was not released by CD8^+^ T cells of NP-mice either upon Ag85B stimulation or following 4-1BB ligation. Fourth, the inability of an IDO inhibitor to counteract 4-1BB regulatory effects indicate that IDO accumulation in macrophages and dendritic cells - a mechanism which suppress CD4^+^ T cell responses in experimental uveoretinitis and rheumatoid arthritis through release of IFN-γ by 4-1BB-activated T cells [@pone.0011019-Seo1], [@pone.0011019-Choi1] - was not activated by Ag85B-specific CD8^+^ T cells and was not involved in control of Ag85B-induced IFN-γ response. The adoptive transfer experiments clearly demonstrated that *in vivo* treatment with anti-4-1BB mAb inhibited antigen-mediated activation, IFN-γ production and expansion of Ag85B-specific CD4^+^ T cells of NP-mice. Although the cellular and molecular mechanisms operating *in vivo* remain to be elucidated, an at least partial involvement of CD8^+^ T cells is suggested by the antigen-mediated expression of 4-1BB on CD8^+^ T cells, and by their proliferation following 4-1BB ligation *in vivo*. Interestingly, it has been recently reported that 4-1BB ligation *in vivo* expanded a subset of CD8^+^ T cells involved in suppression of the recall response to staphylococcal enterotoxin A-specific CD4^+^ T cells [@pone.0011019-Kim1]. Besides inhibition of IFN-γ production, agonist anti-4-1BB mAb treatment of splenocytes of NP-mice induced several changes in Ag85B-specific CD4^+^ T cells consistent with reduced cellular activation. Cell proliferation was lowered, expression of activation marker such as CD38, CD25, CD44 and CD233 was down-modulated, and secretion of MIP-1β and IL-10 was also decreased. The 4-1BB-mediated de-activation of Ag85B-specific CD4^+^ T cells occurred via CD8^+^ T suppressor cells only for IFN-γ modulation while inhibition of IL-10 and MIP-1β secretion was attributable to a direct triggering of 4-1BB on surface of CD4^+^ T cells. Altogether these data indicate that in NP-mice, Ag85B-responding CD4^+^ T cells are a non-homogeneous population or they are regulated by distinct pathways which are differentially affected by 4-1BB ligation. Thus, in mycobacterial antigen-specific CD4^+^ T cells, 4-1BB treatment reduced the release of excessive and dangerous IFN-γ, inhibited secretion of IL-10 and MIP-1β, which are often associated with TB susceptibility [@pone.0011019-Boussiotis1], [@pone.0011019-Jamieson1] while preserved the low production of IL-2 and TNF-α, two cytokines which are essential for granuloma maintenance [@pone.0011019-Flynn1], [@pone.0011019-Bean1] and are expressed, together with IFN-γ, in polyfunctional T cells associated with protection from MTB infection [@pone.0011019-Aagard1]. Overall, the novel regulatory effects shown by agonistic 4-1BB ligands on modulation of potentially dangerous responses of CD4^+^ T cells specific for Ag85B - an antigen actively secreted by replicating MTB and proposed as a candidate sub-unit TB vaccine - open novel strategies for intervention in TB pathology and vaccination through T-cell co-stimulatory-based molecular targeting. Although the role played by 4-1BB *in vivo* during MTB infection needs further investigations, this receptor could be effectively involved in the immunological responses against mycobacterial infection since expression of CD137, the human counterpart of 4-1BB, was found on a sub-population of the T-lymphocytes infiltrating lung tuberculous granulomas in patients with TB [@pone.0011019-Boussaud1]. Materials and Methods {#s4} ===================== Ethics Statement {#s4a} ---------------- The handling of mice were conducted in accordance with the regulations set forward by the institutional animal care committee of the Italian Ministry of Health, and in compliance with European Community Directive 86/609 and the U.S. Association for Laboratory Animal Care recommendations for the care and use of laboratory animals. Immunizations {#s4b} ------------- C57BL/6 female mice were supplied as specific pathogen-free mice by Harlan (Udine, Italy) and were maintained in specific-pathogen-free conditions. Mice, 6 to 8 week old, were immunized. Fifty µg of plasmid Ag85B-encoding DNA, prepared as previously reported [@pone.0011019-Palma1], was injected i.m. in 50 µl PBS into the hind leg. On the dorsum of the mice, 10 µg of recombinant Ag85B protein, prepared as previously reported [@pone.0011019-Palma1], was co-administered s.c. Recombinant Ag85B protein was free of physiologically significant levels of endotoxin, as measured by LAL system. Mice were immunized twice with Ag85B-encoding DNA at 2-week intervals (P-mice). Some DNA-primed-mice were boosted with Ag85B protein twice at 2-week intervals (NP-mice). Four weeks after the last immunization, spleens were harvested for immunological studies. Spleen cell preparation and separation of lymphocytes {#s4c} ----------------------------------------------------- Single cell suspensions from pooled spleens (4 mice per group, 4 separate experiments) were applied to Falcon 2360 cell strainers (BD Discovery Labware, San Diego, CA), centrifuged, separated into aliquots and frozen. From spleen cell suspension, isolated CD4^+^ or CD8^+^ T cells and the corresponding CD4- or CD8-depleted spleen cell populations, were purified by negative selection using specific mouse CD4^+^ or CD8^+^ T cell isolation kits (Miltenyi Biotec Inc. Auburn, CA), in accordance with manufacturer\'s instructions. The purity of isolated CD4^+^ or CD8^+^ T cells ranged between 94% and 96% as estimated by FACS analysis. Spleen cell cultures {#s4d} -------------------- Cells isolated from spleens of immunized mice, either unfractionated or CD4- and CD8-depleted, were cultured at a density 1.4×10^6^cells/ml in 96-well plates for cell proliferation and cytokine detection or in 24-well plates for cytokine detection and flow cytometry analysis in RPMI-1640 supplemented with 10% heat-inactivated FBS, 2 mM L-glutamine, 10 mM HEPES buffer, 50µM 2-β-ME, 50 U/ml penicillin and 50 µg/ml streptomycin (complete RPMI, cRPMI). Cells were stimulated with 5 µg/ml Ag85B protein or 0.1 µg/ml agonistic anti-mouse CD3ε mAb (BD Pharmingen, San Jose, CA) in the presence or absence of 5 µg/ml agonistic anti-mouse 4-1BB/TNFRSF9/CD137 mAb (clone 158321) (R&D System, Inc. Minneapolis, MN) or rat IgG2a isotype control Ab (clone 544479) (R&D System, Inc). In some culture, the IDO-inhibitor, 1-methyltryptophan (1-MT) (Sigma-Aldrich, Saint Louis, MO) 400 µM or neutralizing anti-TGFβ1 mAb (clone 9016) (R&D System, Inc.) were also added. In other experiments, untouched purified CD4^+^ or CD8^+^ T cells were magnetically isolated from the specific mouse CD4^+^ or CD8^+^ T cell isolation kits, (Miltenyi Biotec), in accordance with manufacturer\'s instructions. Purified untouched CD4^+^ T cells (5×10^4^ cells/well) or purified untouched CD8^+^ T cells (3×10^4^ cells/well) were cultured on a feeder of spleen cells recovered from naïve unvaccinated mice and depleted of CD3^+^ T lymphocyte by a magnetic separation through a PE-isolation kit (Miltenyi Biotec) (1.5×10^5^ cells/well). Co-cultures of purified untouched CD4^+^ and CD8^+^ T cells were performed at a CD4/CD8 ratio of 1∶0.6, which reflected the CD4/CD8 ratio observed by FACS analysis in fresh splenocytes. Cells were stimulated with 5 µg/ml Ag85B protein with or without 5 µg/ml agonistic anti-mouse 4-1BB mAb or 5 µg/ml rat IgG2a control Ab. Cytokine detection in culture supernatants {#s4e} ------------------------------------------ Supernatants of cultured cells were harvested at 2, 3, 4 and 5 days and assayed for IFN-γ, IL-10, IL-4, IL-2, IL-6, IL-17, TNF-α, TGF-β, MIP-1β and XCL-1 by specific quantitative sandwich ELISA (mouse Quantikine, R&D System, Inc. or RayBiotech, Inc. Norcross, GA), in accordance with manufacturer\'s instructions. Flow cytometric analysis {#s4f} ------------------------ Fresh spleen cells or cells cultured in 24 well plates for 4 or 6 days, were washed with 1% BSA, 0.1% NaN~3~ in PBS 1× (FACS buffer), and stained for 20 min at 4°C with the following mAbs: PE-Cy5 anti-mouse CD4, PerCPCy5.5 anti-mouse CD4, PE-Cy7 anti-mouse CD8 (eBioscience, San Diego CA), FITC anti-mouse CD11c (Serotec LTD, Oxford, UK), PerCP anti-mouse CD8, FITC anti-mouse CD3, PE anti-mouse 4-1BB, FITC anti-mouse CD38, FITC anti-mouse CD25, PE anti-mouse CD233, FITC anti-mouse CD44, PE anti-mouse CD62L and FITC anti-mouse CD45.2 (BD Biosciences Pharmingen). Cells were washed and analyzed on a FACScan flow cytometer (Beckton Dickinson Immunocytometry System, San Jose, CA) with the Cell Quest software program. Staining of samples with isotype controls was used as a reference to determine positive and negative populations. For intracellular cytokine staining, spleen cells were treated overnight with 10 µg/ml of brefeldin A (Sigma-Aldrich), an inhibitor of protein transport from endoplasmic reticulum to the Golgi apparatus. Then cells were stained for 20 min at 4°C with PE-Cy5 anti-mouse CD4, PerCPCy5.5 anti-mouse CD4, FITC anti-mouse CD45.2 or PercPE anti-mouse CD8, washed and fixed for 5 min with 4% paraformaldehyde. Cells were washed and stained with FITC- anti-mouse IFN-γ, PE-anti-mouse IFN-γ, PE anti-mouse IL-10 or isotype controls (BD Biosciences Pharmingen) in FACS buffer with 0.5% saponin for 45 min at R.T. Cells were then washed in FACS buffer with 0.1% saponin and subsequently with FACS buffer with 0.01% saponin and then analyzed on a FACScan flow cytometer with the Cell Quest software program. Cell proliferation by CFSE staining {#s4g} ----------------------------------- Spleen cells (10^7^ cells/ml) were stained with 5(6)-Carboxyfluorescein diacetate N-succinimidyl ester (CFSE) (Invitrogen Life Technologies) at 1 µM in PBS 1%FBS for 10 min at 37°C in the dark. Cells were washed and cultured in 96-well plates, as previously described, for 4 days. After the incubation time, cells were washed with FACS buffer, and stained for 20 min at 4°C with PE anti-mouse CD4, PerCP anti-mouse CD8, APC anti-mouse CD45.2 (BD Biosciences Pharmingen) or the isotype controls. Cells were washed and analyzed on a FACScan flow cytometer with the Cell Quest or DIVA software programs. Adoptive transfer experiments {#s4h} ----------------------------- 10^7^ splenocytes from NP-mice (about 1.53×10^6^ CD4^+^T cells and 1.16×10^6^ CD8^+^T cells; these T cells are CD45.2^+^) were adoptively transferred intravenously via caudal vein injection into C57BL/6Ly5.1 recipient mice (same strain of mice expressing CD45.1 instead of CD45.2 on their leukocytes). On postinjection (p.i.) day 1, mice (3 mice each group) were inoculated s.c. with 10 µg of recombinant Ag85B protein and then i.p. with 150 µg agonistic anti-mouse 4-1BB/TNFRSF9/CD137 mAb (clone 158321) or with 150 µg rat IgG2a isotype control mAb (clone 544479). On p.i. day 3, mice were again injected with 150 µg agonistic anti-mouse 4-1BB mAb or with 150 µg rat IgG2a isotype control Ab. As controls, a group of mice adoptively transferred with splenocytes from NP-mice was treated with neither antigen Ag85B nor mAbs, and two groups of mice, not inoculated with splenocytes from NP-mice, received only the treatment with Ag85B and mAbs as descrived above. On p.i. day 10, all mice were sacrificed and spleens were harvested for immunological studies. We thank Maria Laura Gennaro for helpful discussion. We are grateful to Federico Giannoni and Lara Brunori, for providing the plasmid DNA-85B, to Maria Elena Laguardia for Ag85B protein purification, and to Giovanni Piccaro and Antonio Di Virgilio for mouse housing and technical assistance. **Competing Interests:**The authors have declared that no competing interests exist. **Funding:**This work was supported by the Ministero della Salute (Italy) grant H51 and European Commission grant LSHP-CT-2003-503240, MUVAPRED Project. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. [^1]: Conceived and designed the experiments: CP. Performed the experiments: CP SV. Analyzed the data: CP SV AC. Contributed reagents/materials/analysis tools: CP AC. Wrote the paper: CP AC.

Expansion and activation of human natural killer cells for autologous therapy. Ex vivo depletion of lymphocytes has been used to decrease the incidence of acute graft-versus-host disease following the allogeneic bone marrow transplantation. Many depletion techniques also remove natural killer (NK) cells that may mediate a beneficial graft-versus-leukemia effect. In this paper we review our studies on ex vivo expansion and activation of NK cells for use in the therapy of chronic myelogenous leukemia.

New British car maker debuts at Frankfurt Motor Show It’s not often that a new car maker joins the motor industry elite, let alone a new British car maker. But, there’s a new British name over one of the stands at the Frankfurt Motor Show. Eterniti (sounds a little like Infiniti the up-market Nissan brand, doesn’t it?) plans to launch next year with the Hemera – a prototype of which will be on the Frankfurt stand. Eterniti say this is the “world’s first Super-SUV”. Based on the Volkswagen Group’s large SUV platform (as used for the Touareg and the Porsche Cayenne), the company says the looks will be very much as the prototype, but the final version will have different LED headlights. The body panels will all be unique and made from carbon composite and the interior is “almost 100% unique” to Eterniti. The new company is backed by international investors and based in London where there will be a showroom and workshop where the bespoke models will be hand build “to any specification required”. The company intends to replicate these facilities in other world cities in the future. The target for the Eterniti Hemera is “wealthy car buyers, particularly in major cities, global wealth centres and emerging markets ranging from London to the Asia-Pacific region”. The business aims to help satisfy a growing demand for additional luxury and personalised specifications among wealthy car buyers, particularly in major cities, global wealth centres and emerging markets ranging from London to the Asia-Pacific region. Eterniti claim that they are the first car maker to enter the luxury SUV market, for which Bentley, Maserati and Aston Martin (with their Lagonda brand) are known to be considering new models. The Eterniti engineering team is led by Alastair Macqueen, who was responsible for the development of the Jaguar XJ220 supercar to production and has won the Le Mans 24-Hour race three times, with both Jaguar and Bentley. Former Formula 1 driver Johnny Herbert will have an active involvement in developing Eterniti vehicles as well as serving as an ambassador for the brand. Plans for another SUV and a high-performance car which could form the basis of a racing programme are currently being developed.

I think we have to remember that it took time for Bergeron to come back from his first concussion. I think that we all hoped that Savard would come back to the level he once was as well, and hoped it was just more time needed. As Amich indicated, he seemed to be moving in the right direction. I agree with David, there is no way we go get Kelley if Savard is healthy. The Bruins have been good about checking their egos at the door, and if Marc Savard’s the third line center he would accept his role. For anyone bashing Savard’s play this past season, where you not watching the games in January? About a week or so before that Avs game he was looking a lot more like the old Savard. He had 4 pts. in 5 games, he was making some great plays, he had a +3 game. . . The projection from others who’ve played in the NHL was it would take him 2-3 mos. to get back to his old self, & it looked like he was on the cusp of being there. We probably don’t go out and get kelley if Savvy is healthy, but I still think the peverly move happens. Even if Savard is downgraded to a 3rd line center, he would still have been better than Kelley, who I won’t be sad to see go. Savard showed he was clutch even after receiving a concussion, so even with abbreviated (for him) production and playing time, it is unlikely that he would have hurt the teams chances in any way. Before his concussion, he was occasionally used on the PK, speaking to the fact that he could play defense, and was still really just coming back when he was concussed again. Also, you have to remember that while his production was poor for his standards, he was on a line with Horton who at the time was in a slump for the ages, which does not help one’s own points total. It is possible that we don’t go and get peverly if Savvy is healthy, but then again, our power play probably wouldn’t have been a steaming pile of shit either. All and all, what happened happened. And we con the cup, so I can’t complain too much. Frank, your glass is definitely half full. I think missing Savard forced the bruins to pick up certain players they might not have otherwise and they definitely juggled the lines as a result. Winning the cup is next to impossible with team parity where it is right now. For all the necessary stars to line up to win a cup, I have to believe Marc Savard not being in the lineup was a big part of it. I’m not congratulating anyone for a headshot though. It’s sad but Savard is a big chunk of change for only one fragile man. bruins16: stop living in the pre-Cooke hit era. No doubt the Bruins would’ve won the Cup this year with the ’08 or ’09 version of Savvy. No argument. But that’s not the case: it was the ineffective ’11 post-Cooke hit version this year. That’s all I’m saying. Stop living in the past. Well said nifty. Believe me: I would truly prefer to think otherwise. David: it might not be ‘fair’ to Savvy as a human being but it’s the hurtful reality. Kenisha: you really think the Bruins would’ve missed the playoffs without the 10pts of Savvy (even if they finished 10pts ahead of the Rangers…)? tom: Peverley & Kelly are 3rd liners who don’t even play on the PP. I hope your expectations are way higher for a 1st liner who’s on the 1st PP unit… No way the league should bend the rules and put his name on the cup. That is why they have rules as I am sure there are other players in the past that feel they should have had their name on the cup but didn’t. But of course with Colon Campbell having some say, anything to keep the Bruins happy will get done and Savard will have his name on the cup. What a joke. Frank, You do know you need to be in the top 8 in the Eastern Conference to make playoffs, right? Without Savard in Boston, the Bruins would have NOT been making playoffs. Thus, not being able to win the Cup. Savard allowed the Bruins to gain the confidence and determination that they needed to win the Cup. Savard allowed the Bruins to have the willingness and the DRIVE to win. He gave them a reason to not want to back down. Hockey isn’t all about skill, if it were, Ovechkin would have a ring; but it’s also about inspiration. Ever heard of “Do it for Horton” or “Do it for Savvy”? Frank, It’s not at all fair to say that we would have lost the cup if Savard were healthy. In the games he played, he was obviously still affected by the Cooke concussion, and also not in the best of shape as a result. Had he had more time, he would have been better or asked to take some time off because of the concussion, and I do recall Julien benching him for a few periods because of ineffectiveness, so the comments are pretty far off base. Also Karl, I think Wheeler’s name will be on it, as well as Stuart’s because each played 41+ games in a Bruins sweater last year. What are you talking about Mike? If Savard would have been healthy, no doubt the Bruins would NOT have won the Cup. Claude would’ve felt the obligation of giving him premium ice time despite his obvious ineffectiveness. 2 goals & 8 assists in 25 games for a player who doesn’t hit & is the worst defensive center of the team can’t be good… We should all thank Matt Hunwick for causing his injury & giving us our 1st Cup in 39 years. Though, I’m totally in favor of making an exception & engrave his name on the Cup. He would help his case a lot by announcing his retirement before the decision is taken. This is a no-brainer. The Bruins certainly wouldn’t have evolved into a Stanley Cup winner without having #91 in the fold. Even though he didn’t participate in the 2011 playoff run, he’s been an integral part of the team, he did play 25 games during the year and, as TCL mentions, would certainly have played a critical role in the Cup run, had he been healthy enough to participate. I hope the league does the right thing and adds his name to history.

Yum reports 2Q profit dip, but raises outlook Yum! Brands Inc. reported Tuesday a dip in second-quarter profit, but raised its full-year earnings forecast on what it called a strong first half of the year. For the quarter ended June 12, Yum earned $286 million, or 59 cents per share, compared with earnings of $303 million, or 63 cents per share, in the same quarter a year ago. In the second quarter of last year, Yum had booked a one-time gain of $68 million from the purchase of additional ownership in its KFC business in China. ... Register to view this article It’s free but we need to know a little about you to continually improve our content. Why Register? Registering allows you to unlock a portion of our premium online content. You can access more in-depth stories and analysis, as well as news not found on any other website or any other media outlet. You also get free eNewsletters, blogs, real-time polls, archives and more. Attention Print Subscribers: While you have already been granted free access to NRN we ask that you register now. We promise it will only take a few minutes!

Here is a text from Flame with his big fish. I will wait to comment until Flame has added his wisdom. I only hope the pictures come through. "Caught this yesterday 3.5 pounds at only 16.5 inches. Relative weight came in at 140%. Pure Florida lm pellet trained from Todd Overtons. This fish was only 4inches long only 2 years ago this month!! Me proud paw paw!! Thanks for your help. We wanted them to know this pond is in east Texas. Pat says everything is bigger in Texas!!" Thank you CMM for helping me with posting the photos. I was hoping I had a 3 pound lmb in my 2 acre pond. I only stocked 50 fish and the largest up till now had only been 2 and 1/2 pound. Boy did I get a surprise!! Caught it on a plum wacky worm and man did it fight!! I was by myself so did the best I could do with taking photos. _________________________ Dear Alcohol, We had a deal where you would make me funnier, smarter, and a better dancer... I saw the video... We need to talk. Nice lmb Flame!! I am betting you will continue to get good growth on your lmb. Mr Lusk said after they reach 3 lbs they will rule the pond. _________________________ Do not judge me by the politicians in my City, State or Federal Government. Thank The Good Lord the government in Washington DC gets little done.Outlawing guns will make a lot of us down here in the SouthOutlaws and proud of it I'm happy for that old lmb fisherman I know of his cnbg population and I bet he see's some great lmb growth this next 12 months. _________________________ Do not judge me by the politicians in my City, State or Federal Government. Thank The Good Lord the government in Washington DC gets little done.Outlawing guns will make a lot of us down here in the SouthOutlaws and proud of it Thanks again CMM for posting that picture. This is the first one I caught while my wife was fishing with me so I got her to take the picture. I'm just blown away by the growth and so proud!! Now if I can just get my wife to hang one of them bad boys!! She had one on the other day but it pulled off and she didn't get to see it. Thanks everyone for the kind words. _________________________ Dear Alcohol, We had a deal where you would make me funnier, smarter, and a better dancer... I saw the video... We need to talk. Another nice one Flame For me I am not fishing the lmb or hsb this time of year because when fishing a couple of weeks back I lost half of the lmb I had caught. Found them dead and floating a couple of days after the fishing. I am not sure if it's just my pond or if it happens in most every E Texas pond this time of year. I also read a while back when catch and release fishing lmb in Texas lakes during the summer there was a high mortality rate on those fish released. It was because the fish did not float up till a couple or three days after their release, the fisherman did not realize his released fish were dying. Seems like fishing bigger lakes the fish were not seen by the fisherman due to it being a few days after the fish was released. I do know that when the fish are deeper in the summer they have to or need their bladder de-aired by using a syringe or they will die. But I am not talking of deeper fish in my 10 to 12 foot water depths. Is it because my fish are already stressed before they are caught? Edited by TGW1 (06/12/1808:14 AM) _________________________ Do not judge me by the politicians in my City, State or Federal Government. Thank The Good Lord the government in Washington DC gets little done.Outlawing guns will make a lot of us down here in the SouthOutlaws and proud of it Tracy, a couple of them did take quite a few minutes to swim off out of my hand. They were the first 2. I did "fight" those fish longer than needed but really wanted to land them for weights and measures. I was by myself and I know I stressed them more than I should have. This last one I had the wife to help with getting scales and measuring and photos so was able shorten the landing and weighing time a lot. This one swam off as soon as it entered back into the water. But I do agree,those first ones told me it is getting too hot. But on another note, those pure Florida lmb that size fight so hard you gotta wear em down some before landing!! Can't help it ...I'm having FUN!! _________________________ Dear Alcohol, We had a deal where you would make me funnier, smarter, and a better dancer... I saw the video... We need to talk. keep having fun Flame!! just wanted to tell my exp. with loss of lmb after release. I hate losing 4lb lmb out of my pond. _________________________ Do not judge me by the politicians in my City, State or Federal Government. Thank The Good Lord the government in Washington DC gets little done.Outlawing guns will make a lot of us down here in the SouthOutlaws and proud of it

Havistin Havistin (, also Romanized as Hāvīstīn; also known as Havestīn and Havīstūn) is a village in Harzandat-e Sharqi Rural District, in the Central District of Marand County, East Azerbaijan Province, Iran. At the 2006 census, its population was 330, in 90 families. References Category:Populated places in Marand County

[Role of matrix glycoproteins during chondrocyte differentiation (author's transl)]. The importance of cell-matrix interactions during terminal chondrocyte differentiation is to-day accepted by all authors. Na-hyaluronate micro-injections induce an increase of perichondral thickness and maintain mesenchymal aspect in chondrocyte. Cytoplasmic organelles show anomalies which may be bound to disturbance in cellular metabolism. An important decrease of matrix-constituents is observed, a fact showing that high hyaluronate concentration interferes with chondroitin-sulfate synthesis or secretion and collagen organization.

October 16, 2010 - Susan Ivey announced that she would retire from Reynolds American Inc. after nearly 30 years in the tobacco industry. Ivey, 51, said she would retire as chairwoman on October 31, then as chief executive, president and member of the board of directors on February 28. The Reynolds board named Daniel “Daan” Delen, 44, to succeed her as president, chief executive and board member, effective January 1, 2011. The board elected Thomas Wajnert, 67, as a non-executive chairman, effective November 1, 2010 . He has been the board’s lead director since 2008. The company said that Ivey plans to move to Florida to be near family and to spend more time traveling with her husband, Russell Cameron, who is already retired. From a business perspective, the news was surprising since Ivey appeared to be in the prime of her career. Ivey, has been chairman, president and CEO of Reynolds American since 2006. She served as RAI’s president and CEO and a member of the board of directors between 2004 and 2006. From 2001 to 2004, Ivey was president and CEO of Brown & Williamson Tobacco Corp. Brown & Williamson and R.J. Reynolds Tobacco Co. combined their U.S. businessesin 2004. Ivey will enter retirement as one of the nation’s mostly highly paid female executives. For example, her total compensation for 2009 was $16.2 million. Locally, even though Reynolds had an 18 percent loss of jobs, or about 600, during her years as chief executive, she slowed the pace of cuts, compared with the loss of 12,100 jobs from 1983 to 2004. Analysts said they were somewhat surprised by Ms. Ivey's decision because she is a young CEO. But they said she was leaving Reynolds—the second-largest U.S. tobacco maker by sales after Altria Group Inc.'s Philip Morris USA—in good financial shape in an industry that faces declining cigarette sales and continuing scrutiny from regulators and anti-smoking groups.Graphic by Nicolas Weir, ofthe Winston-Salem Journal.. "Susan has turned Reynolds into a very good No. 2," said UBS analyst Nik Modi. "She's young, but she's had a heck of a career." Ms. Ivey, who was the first woman to run a major tobacco company, has transformed the maker of Camel and Pall Mall cigarettes by slashing its portfolio of cigarette products and reducing production costs. In 2006, she spearheaded the acquisition of Conwood Co., the maker of Grizzly moist snuff, a successful entry in the growing smokeless-tobacco category. (Reynolds American Inc. acquires Conwood..) After ongoing evaluation, the board also decided to split the roles of chairman of the board and chief executive officer, in line with best-practice trends in corporate governance. Thomas C. Wajnert was elected non-executive chairman of RAI’s board, effective Nov. 1, 2010. Wajnert has served as the board’s lead director since 2008, serves as the chair of the board’s compensation and leadership development committee, and is a member of the corporate governance and nominating committee. He chaired the board’s audit and finance committee from 2004 to 2009. October 15, 2010 - A ruling Democratic Party of Japan (DPJ) committee recommended not raising cigarette taxes next year after this year’s 40 percent increase because the government should examine the impact on tobacco farmers and manufacturers. Background: “We should aim to boost the tax again next fiscal year,” Yoko Komiyama, a member of Japan’s ruling party and a leader of a group of about 70 lawmakers seeking higher cigarette levies, said in an interview in Tokyo. “The planned increase next month is just the first step and isn’t nearly enough.” (Japanese lawmakers want to triple cigarette prices..) Komiyama said the aim should be a price of 1,000 yen (11.68 USD) a pack, which would bring Japan in line with some European countries. The government could raise the price by about 100 yen (1.17 USD) every year to allow the nation’s tobacco leaf farmers enough time to switch to other crops, Komiyama said. “A consensus is growing in the government that we have to do more to discourage things like smoking that are harmful to health,” she said in an interview on Sept. 9. Japan Tobacco, which controls 65 percent of the domestic market, “strongly opposes” a further tax increase until the affect of the October price change is assessed, the company said in a statement on Aug. 31. The DPJ’s fiscal panel released its tax reform proposal today, October 13th calling for a “cautious” approach in raising tobacco taxes after this month’s record increase lifted the price of a pack of cigarettes by one-third to about 410 yen ($5).The recommendations may complicate Prime Minister Naoto Kan’s efforts to increase government revenue to pay for social welfare programs. The administration is seeking to ensure economic growth while reining in the world’s largest public debt. Japan is the fourth-largest market by volume for the world’s tobacco makers, after China, the U.S. and Russia, according to a report from U.K.-based market researcher ERC Group. The DPJ panel said Japan should consider raising cigarette taxes to the levels of Western countries in the future for health reasons. October 15, 2010 - Omega the chimpanzee spends his days in a cage, smoking cigarette ends. Originally trained to perform tricks at a restaurant, he soon became too strong – and too unhappy – for his role, and was sent to live in a zoo in southern Lebanon. Now 12 years old, he has not seen another chimp, or climbed a tree, in 10 years.Animals Lebanon – a non-governmental organization (NGO) that works to educate for “A better Lebanon, one animal at a time” – has taken up Omega’s cause. The owner finally agreed to close thesmall zoo – which was damaged by the 2006 summer war with Israel – but the responsibility for re-homing the animals now rests with Animals Lebanon. Jason Mier, the NGO’s executive director hopes to find homes for some of the animals, such as the ducks, geese and rabbits locally. It is not so easy with an adult chimpanzee, but Animals Lebanon has now found a suitable home for Omega, over 10,000 kilometers (6,213.7 miles) away, in South America. A dedicated chimpanzee reserve, The Vargem Grande Paulista Sanctuary in Sao Paulo, Brazil, has offered Omega a permanent home. The sanctuary will offer Omega a home as close to his natural environment as possible.Import and export permits have now been issued by the respective governments, and the NGO is in talks with Qatar Airways to organize Omega’s flight to Sao Paulo.They need to find a plane that can carry live animals, and then pay for the cargo – a heavy load considering Omega has to be transported in a metal cage weighing around 200 kilograms. “At $10 or $12 per kilogram, this is quite an expensive flight,” Mier added. The NGO is asking for donations via their website animalslebanon.org to help with Omega’s safe and prompt transit to his new South American home, where he will interact with other chimpanzees for the first time in a decade. October 15, 20010 - Mayor Michael Bloomberg and City Council leaders want to extend the Smoke Free Air Act, a 2002 law that banned smoking from offices, bars, restaurants in an effort to protect people from the harmful effects of secondhand exposure. (New York City officials have announced a plan to ban smoking outdoors. - see background information) The bill introduced by Councilwoman Gale Brewer, a Manhattan Democrat, with the support of the Bloomberg administration, would ban smoking in the public parks, playgrounds, beaches and pedestrian plazas. A New York City Council public hearing on a proposed smoking ban took place on Thursday, afterrnoon, October 14th touching on issues such as civil liberties, public health, big government and litter. Testimony ranged from people like David Goerlitz, the former “Winston Man” who, in a press conference before the hearing, said smokers are treated like “lepers and second-class citizens,” to Joe Applebaum, a Brooklynite who equated second-hand smoke with rat poison and said smokers have “no consideration for their fellow man.” Council member Peter F. Vallone, Jr., a Queens Democrat, who described himself as an “anti-smoking advocate,” has introduced a bill requiring that land under the jurisdiction of the City Department of Parks and Recreation that is larger than two acres must have a designated smoking area equal to at least a fifth the size of the property footprint.Backers of a ban say that even brief exposure to secondhand smoke can pose health risks. Dr. Farley, New York City Health Commissioner, and other proponents, including the Coalition for a Smoke-Free City, an advocacy group, and organizations like the American Cancer Society of New York and New Jersey, also said smoking in parks sets a poor example for youth. "There is no safe level of second-hand smoke. Not inside, not outside, not anywhere,” said Dr. Maureen Killackey, Chief Medical Officer of the American Cancer Society. In addition to the health issue, city officials said they are also concerned about litter. They said they’ve found that 75 percent of the debris on beaches, 45 percent on playgrounds and 33 percent in parks were cigarette butts or cigarette packaging. City Council sources told CBS 2 Marcia Kramer reporter the bill was being fast tracked and a vote is expected within a month. October 15, 2010 - A groundswell of renters and owners are demanding smoke-free housing in multiunit dwellings, the final frontier in a fight that brought bans in workplaces, on airplanes, in restaurants and bars, on patios, near doorways, in prisons, and, just last month, in parks. But the latest debate literally hits home, and pits tobacco users, who argue they’re consuming a legal product, against people who don’t want harmful smoke drifting into their suites through vents, hallways and windows. “It’s one of the next major health issues – I think in part because of the number of people who are now living in apartments and condos,” said Sharon Hammond, who helped launch smokefreehousingbc.ca, a website that offers advice to landlords, tenants and condominium owners.She rejects the idea that the push to restrict smoking in private residences is a matter of conflicting rights, saying it is simply an extension of the logic behind ever-tighter government restrictions on tobacco use at work and in public places. “People are also getting used to not smelling smoke everywhere. Then they get home and think, ‘If I don’t have to smell it when I’m out, why do I have to smell it in my own home?’ ” For the most part, governments are remaining on the sidelines, but a handful of Canadian municipalities are already curbing smoking in shared dwellings. This year, Waterloo, Ont., banned smoking for all new residents in public housing, and began offering counselling and nicotine replacement for those who want to quit. St. John’s has adopted a similar policy, and several other jurisdictions are looking at following suit. In the United States, the limits are more widespread. In 2000, just two public housing authorities had smoke-free policies; now that number has risen to more than 200. One municipality in California has put an outright ban on smoking in all multistory, multiunit residences. Some states, including Oregon, require building managers to disclose whether smoking is allowed and where. In Canada, there are no bans on smoking in market-rate condos and rental apartments, but an increasing number of developers are looking at building smoke-free complexes – especially those seeking the coveted green LEED certification, (LEED stands for Leadership in Energy and Environmental Design and is a rating system performed by a third party on how green your building is.) which puts strict limits on smoking. Ms. Hammond and other opponents of smoking say such voluntary steps, along with a disclosure law similar to that of Oregon, are their preferred short-term route to smoke-free living. The anti-smoking lobby is also hoping for regulations making new condos smoke-free unless owners vote otherwise. Although 85 percent of people in British Columbia (B.C.) don’t smoke, and the province has more than 725,000 residential strata lots, fewer than a dozen condo complexes have outlawed smoking. Still, some are concerned about the increasing calls for smoke-free apartments. BC Civil Liberties Association president Robert Holmes says the rights of smokers must be carefully weighed against the rights of non-smokers. Even though secondhand smoke has been deemed harmful, the degree of harm must be considered, since urban life is full of dangers, from car exhaust to cologne. Tony Gioventu says it’s not that simple. As executive director of the Condominium Home Owners Association of BC, he argues that when people buy into a condo development, they are signing away their right to do whatever they want. U.S. District Court (for the Western District of New York) Judge Richard J. Arcara on Thursday, October 14th extended indefinitely his ban on state taxation of cigarette sales by the Seneca Nation and other Indian tribes. But it was not a complete victory for the Senecas and their supporters. In one of two rulings, Arcara extended a temporary restraining order that prevents the state from collecting taxes on cigarette sales by Indian-owned businesses to non-Indians. The temporary order, which the judge initially issued during the summer, had been scheduled to end today. judge initially issued during the summer, had been scheduled to end today. Arcara said Indian tribes throughout the state would "suffer irreparable injury" if he did not extend the temporary order. He said thousands of smoke shop workers would be likely to lose jobs if the state taxation begins now. The judge also voiced concerns about public safety if he does not extend the order. He said he is concerned about the threat of violence by Indian protesters if the taxation begins. Arcara also filed a second order in which he turned down some of the arguments that Indian tribes have presented to fight off taxation. He said lawyers for the Seneca Nation and the Cayuga Indians have "failed to demonstrate a likelihood of success" on their claims that the state taxation plan is unconstitutional. Previous decisions in federal courts "make clear that the [Indian tribes'] right to tribal self-government is not impeded by New York's decision to impose a tax-collection duty on sales by Indian retailers to non-members," Arcara wrote, "even if that decision carries with it the inevitable consequence that the [tribes'] coffers will suffer as a result of lost cigarette sales." While Seneca Nation leaders and protesters publicly insist that the state's taxation efforts violate Indian treaties, that argument has not been advanced by the tribe's lawyers in the litigation before him, Arcara said. In its legal arguments, the Seneca Nation "expressly acknowledges that, as a general principle, New York State has the authority to require reservation retailers to collect excise taxes on sales to non-Indians," Arcara wrote. "This point is significant. New York estimates that of the 10 million cartons [of cigarettes] sold last year by [Seneca] retailers, less than 70,000 were purchased by Seneca Nation members for their own personal consumption. Under [previous federal court rulings], the vast majority of sales made by reservation retailers are taxable," the judge wrote. Arcara said he has already been informed by Seneca and Cayuga lawyers that they would appeal his ruling. In his view, he wrote, the best course of action is to extend his temporary restraining order while a higher court hears legal arguments on the controversial case.The lawsuit filed by the Senecas and Cayugas is not the only case on the issue pending before Arcara; he has similar cases filed by the St. Regis Mohawk tribe and the Unkechauge tribe. Another case filed in Utica by the Oneida Indians is expected to be transferred to him. (New York State - tribal lawsuits may be merged within federal court..) The bottom line, according to many court observers, is that the state could eventually succeed in its efforts to tax Indian cigarette sales, but the multiple legal challenges could block taxation for months or even years. Pfizer said it supplied the prescription drug for about 70,000 people a month until August, but the number jumped to about 170,000 in September and the company had already supplied about 80,000 people in the first six days of October. The demand has grown far bigger than the drugmaker expected and it has asked medical institutions to postpone prescribing the drug to new patients. The company said that while it is working to increase supply, new patients may not be able to try the drug until around early next year. The orally administered tablets block the part of the brain that receives nicotine, making it difficult for recipients to crave cigarettes, according to Pfizer. When one begins the treatment, a small dose is taken once a day. The doses are gradually increased and the medication continues for three months. "While it was beyond my expectation that (demand for the drug) has grown this much with only an increase of about ¥100 in cigarette prices, it is troubling that I can't prescribe it to new patients," said Yoji Hirayama, a doctor at Tokyo Medical University Hospital. October 15, 2010 - Philip Morris International Inc. (PMI) will host a live audio webcast at www.pmi.com on Thursday, October 21, 2010 at 9:00 a.m. ET to discuss 2010 third-quarter results, which will be issued at approximately 7:00 a.m. ET the same day. During the webcast, Hermann Waldemer, Chief Financial Officer, will discuss the company's 2010 third-quarter results and answer questions from the investment community and news media. The webcast will be in a listen-only mode. An archived copy of the webcast will be available until 5:00 p.m. ET on Friday, November 19, 2010 at www.pmi.com. October 14, 2010 - KT&G Corp., Korea’s tobacco monopoly, has recently completed the construction of a plant in Russia to expand its presence in the world’s second-largest tobacco market. The company has invested $100 million in building the plant located on a 103,000 square-meter site near Moscow. Focusing on its flagship slim cigarettes, , the plant will produce 4.6 billion cigarettes annually, KT&G said. It began operating Friday, October 8th. Launched in 2002, Esse is the No.3 best-selling brand in the super-slim category in Russia, making up over 10 percent of the market. Between 2006 and 2009, slim cigarette sales soared 91 percent, accounting for 13 percent of the total cigarettes sold in Russia, according to Euromonitor International, a London-based market research firm. Esse is also making a strong showing in Eastern Europe, ranking third in the super-slim category in Poland and Ukraine.The construction of the Russian plant is the latest in KT&G’s efforts to expand international business and localize overseas operations. Its growing presence abroad has offset declines in domestic demand resulting from health concerns. The world’s sixth-largest tobacco company currently operates plants in Turkey and Iran. Russia consumes nearly 400 billion cigarettes a year with about 44 percent of adults known to smoke. October 14, 2010 - Despite intense opposition from the international tobacco industry, Australia's plan to introduce plain-wrap cigarettes is likely to be followed by other countries, the Health Minister, Nicola Roxon (Nicola.Roxon.MP@aph.gov.au) says. But Ms Roxon said the positive reaction of ministers from other Western countries to the policy showed that the tobacco companies' campaign was based on a false premise. She told the (Sydney Morning) Herald she had encountered "a lot of interest" from health ministers at an Organisation of Economic Co-operation and Development conference (OECD) in Paris last week. A key theme of the conference was measures to reduce health costs at a time of tight government budgets. Ms Roxon said that cutting tobacco use was often raised as a solution during discussions at the conference.Representatives from Ireland, Norway, the European Union and the United States had expressed interest. "From my perspective, the broader the interest there is, the better." The tobacco companies have made clear they will fight this tooth and nail because they think if it gets through here it will be a precedent. Ms. Roxon: "Interestingly a lot of the other countries were very quick to understand that that fight is actually a bit of a false one because a lot of them are interested anyway. It is sort of an inevitable step along a spectrum of taking action against tobacco. What I was heartened by is the potential that other countries would like to move in this direction irrespective of ours, which makes the tobacco company fight with us a little different. People were pretty gob-smacked about the amount of money the tobacco companies spent during the campaign, especially relative to the amount of money the political parties spent." Australia has reduced the proportion of smokers in the population to 15.5 per cent and the death toll from smoking-related diseases has fallen to about 15,000 - down by about 3000 from a decade earlier. Ms Roxon said she expected drafting of the legislation and the Health Department's assessment of the plain packaging policy to start by the end of the year. The anti-tobacco campaigner Professor Simon Chapman, of Sydney University, said it appeared the tobacco industry was scaling down its campaign. There had been suggestions that the industry was to continue with a multimillion-dollar campaign during the football final telecasts, but this did not eventuate. October 14, 2010 - A last-minute rush ahead of a record tobacco tax hike boosted cigarette sales in Japan by 88 percent from a year earlier to 37.4 billion sticks in September, an industry association said Wednesday October 13th. "The sales are expected to decline in October and gradually recover. But it is difficult to predict how much they will recover because the price increases were of an unprecedented scale," he said. Japan - with tax increase in place, fewer smokers buying cigarettes so far..Japan Tobacco (JT), which dominates the country's cigarette market and in which the government has a 50 percent stake, raised prices by around 100 yen (1.2 dollars) for a pack of 20 on October 1 making some brands nearly 40 percent more expensive. Popular brand Mild Seven rose from 300 yen to 410 yen per pack, up 37 percent. October 14, 2010 - A scheme offering hand-outs of £50 (79.97 USD) a month to those trying to quit smoking has seen less than one in three succeed in stubbing out their habit. NHS Tayside's Quit4U and Give It Up for Baby schemes have cost the health board more than £4.5 million (7.2 million USD) over the last five years. However, despite the huge investment, only 29 percent of those who tried to give up while being given the handouts actually kicked smoking. New figures released in a Scottish Parliament answer show £949,000 was spent on smoking cessation in Tayside over the past four years, while a further £790,000 was spent in the region in 2006-7. When it was launched, the TaxPayers' Alliance pressure group described it as an "extremely dubious use of taxpayers' money". Mr Ballard said said: "The cost of treating people for smoking related diseases far outweighs the figures you have for quitting in terms of investment. "NHS Tayside has focused its investment on the areas of Tayside which have the highest populations of smokers and these areas, without exception, are areas of high deprivation. "We are trying to help those in greatest need, so we focus our energies and investment on the most deprived." A report found tobacco smuggling in Wales is a threat to public health as well as a loss in tax revenue to the Government. Smokers questioned over their attitudes to buying illegal and counterfeit tobacco found younger smokers more likely to buy from the illicit market in cigarettes – fuelling addiction in youngsters. More than half (52%) of 16 to 34-year olds admitted buying cheap tobacco compared to 41% of 35 to 54-year-olds, and 36% of the over 55’s. Back on August 18, 2008 ASH Wales survey found similar results from a survey of 100 underage smokers (average age is just 14). (Wales - still easy for youngsters to buy cigarettes..Some 68% of smokers quizzed were more interested in cost than where their cigarettes come from. Tanya Buchanan, chief executive of ASH Wales, said: “Smuggled tobacco is a bigger issue in more socially disadvantaged areas and this evidence further supports that. Smoking prevalence rates are already much higher in these areas and are already creating significant ill health for residents. Unfortunately, this evidence also means that measures to protect children from accessing age restricted tobacco products are being undermined by tobacco smuggling. “It is vital that the Welsh Assembly Government’s tobacco control action plan being drafted addresses tobacco smuggling and that they work closely with the UK government on this issue. “We need a coordinated multi-agency approach to tackle large scale organised smuggling. Enacted July 1, 2007..October 14, 2010 - MP (member of parliament) David Nuttall (NuttallBuryNorth@aol.com) failed in his attempt to permit landlords to decide for themselves whether to allow smoking in their pubs. Mr. Nuttall's bill was defeated by 141 votes to 86. David Nuttall, the new Conservative MP for Bury North, wants to overturn the smoking ban by introducing a Commons Early Day Motion. Nuttall: landlords should be allowed to decide for themselves whether to allow smoking in their pubs. Mr Nuttall said the legislation was partly responsible for the closure of dozens of pubs in his constituency. Mr Nuttall's bill would exempt pubs and social clubs from the ban, allowing landlords and licensees to have dedicated smoking lounges for drinkers complete with smoke filters. "I'm a devout non-smoker and wouldn't want to go back to the days when people were eating a meal and had smoke blown into their faces," said Mr Nuttall. He added: "We do want to add an element of freedom of choice." Hakim Sorimuda Pohan, from the Coalition Against Corruption of the Anti-Tobacco Clause (Kakar) and a delegate to the APACT, said the government should urgently address this matter. He added that it was the first time in the history of the APACT that the conference’s final declaration included a call for a nation to sign the FCTC. The declaration was backed by 700 delegates from 41 countries who attended the conference from October 6-9, 2010. “This is very embarrassing for Indonesia not to have signed the convention, while other Asian countries have already ratified it,” Hakim said. “By failing to sign it, we’re protecting the tobacco industry rather than the health of our own people.”He warned that if Indonesia did not sign the FCTC soon, it would be “excluded from international society, and we don’t want to be an isolated country.” Harley Stanton, the APACT president, said, “We are deeply concerned that in failing to accede to the FCTC, Indonesia is creating a public health disaster.“Indonesia’s tobacco-related death toll is more than 200,000 a year. More than 60 percent of Indonesian men smoke.” He said more and more Indonesian children were also smoking, with a 400 percent increase in smoking among 5 to 9 year olds. Cigarettes sell for about a dollar a packet in Indonesia, where 60 percent of men are estimated to smoke along with a "disturbing" number of children, according to the Asia Pacific Association for the Control of Tobacco (APACT). Finance ministry official Agus Supriyanto told Dow Jones Newswires the excise level was likely to rise five percent on average next year in what is the world's fifth-largest tobacco market. Customs chief Thomas Sugijata said the increase would help meet the country's excise revenue target of 60.7 trillion rupiah (6.8 billion dollars) -- up four trillion rupiah (448 trillion dollars). But he said the final excise rate had still to be decided and was subject to discussion with the industry, which employs millions of people in the archipelago. "We're still discussing the rate increase and will ask for input from industry, but we can promise that the increase will be moderate," Sugijata said. For 2010 the government aimed to reduce cigarette production to 240 billion sticks from 245 billion last year. The current excise level per cigarette ranges from 65 to 320 rupiah (0.007 to 0.035 dollars). October 13, 2010 - The European Commission is preparing to introduce legislation in 2011 to ban smoking in public places right across the union. (The European Commission is the executive body of the European Union. The body is responsible for proposing legislation, implementing decisions, upholding the Union's treaties and the general day-to-day running of the Union.) While partial or total smoking bans have been introduced in many European countries ending patrons' ability to smoke in bars, cafes and other public venues, it is still relatively easy in some states to find a bolt-hole where smokers are welcome, whether due to exceptions to such laws or owners flouting the bans. Health commissioner John Dalli (responsible for Health and Consumer Policy) has said he wants to put a stop to this. Announcing that Brussels is currently preparing a bill to be brought forward next year, he said that exceptions should no longer be tolerated, as the matter "is not only about the health of visitors, but also the employees." In Belgium for example, home to the EU executive, patrons can still light up in cafes so long as the establishment does not serve food, while the Greek health minister last Wednesday admitted the effective failure of its 2009 smoking ban, saying that undercover inspections revealed that eight out of 10 bars openly flouted the law. The EU bill may also attempt to reduce the amount of nicotine and other toxic substances contained in the product. The commission will furthermore try to win agreement on rules making tobacco products no longer visible to customers and make packaging as unattractive as possible. The packets are to be made identical in appearance and to bear colourful warning pictures, such as of diseased lungs, as well as more information on the toxins the product contains. "The more uniform and bland packaging the cigarettes are, the better," said the commissioner. Within the frame work of the provisioned regulation, oriental tobacco blends may be totally banned as the organization aims to stop “input material” in cigarette production. If the regulation is approved in November, Turkish tobacco exporters will be facing a serious threat. Turkey is the chief cultivator and exporter of the type of tobacco known as oriental tobacco, according to the statement. Turkey’s tobacco production constitutes 4 percent of global production of 7 million tonnes, placing Turkey fifth after China, India, United States of America and Brazil. The world cigarette market is in general based on blended cigarettes, which include a certain amount of oriental-type tobacco. About 65 percent of oriental tobacco is produced in Turkey, 25 percent in Greece and 10 percent in Bulgaria and the former Yugoslavia. Production of Virginia and Burley tobacco amounts to little more than 3 percent (8 000 tonne) of total tobacco production in Turkey. It is felt that tobacco growers and workers should be involved at every stage of policy development and implementation. WHO’s decision to present final guidelines for adoption by the 171 member countries of the Framework Convention on Tobacco Control (FCTC) at a meeting to take place in Uruguay in November 2010.The guidelines for FCTC articles 9 and 10, would ban the use of ingredients other than tobacco in cigarette production. These ingredients are necessary in the manufacture of blended tobacco products, composed primarily of Burley and Oriental tobacco, which account for more than half the cigarettes smoked in the world, outside China. Without them, farmers of Burley and Oriental would see demand for their crops disappear. The recommendations for articles 17 and 18 are meant to provide viable crop alternatives to tobacco growing, but fail to present economically feasible options for tobacco farmers. The proposal risks decimating growers’ livelihoods, condemning millions to a life of poverty and crippling the economies of many developing countries - the very same countries the WHO is funded to help.

Afton Street Conservation Reserve The Afton Street Conservation Reserve is situated along the middle reaches of the Maribyrnong river in West Essendon, Melbourne. It is within the City of Moonee Valley and is 17 hectares (42 Acres). History The Conservation park was purchased by the city council in October 2003 for a price of $900,000 (AUD). The site was formerly part of the Defence Explosive Factory Maribyrnong. Planning A master landscape plan was issued for the site and won a number of design awards. The plan emphasises conservation and recreational use of the park. Flora and fauna Two areas of Western Basalt Plains Grassland 25 regionally significant species of indigenous plants potential habitat and occurrence of the nationally significant Striped Legless Lizard areas of valuable native fauna habitat including five ‘birds of prey’ using open grassland areas, small passerine birds including thornbills and silvereyes using the shrub layer and waterbirds using the riparian fringe References Category:Parks in Melbourne

S (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1))) (2 (2 (2 2 3) (3 4 5)) (3 (4 6 7) (5 8)))) (2 (3 (4 (6 9 10) (7 11 12)) (5 (8 13 14) (9 15 16))) (4 (6 (10 17) (11 18)))))) @S (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12) (7 13)))) (1 (2 (4 (8 14) (9 15 16)) (5 (10 17 18) (11 19 20))) (3 (6 (12 21) (13 22)) (7 (14 23 24) (15 25))))) (1 (2 (4 (8 (16 26) (17 27)))))) PP (0 (0 (0 (0 (0 (0 0) (1 1 2)) (1 (2 3 4))))) (1 (1 (1 (2 (3 5 6)) (3 (4 7 8) (5 9 10))) (2 (4 (6 11 12) (7 13 14)) (5 (8 15 16) (9 17 18)))) (2 (3 (6 (10 19 20) (11 21)) (7 (12 22 23) (13 24 25))) (4 (8 (14 26 27) (15 28 29)) (9 (16 30 31) (17 32 33)))))) IN (0 (0 (0 (0 (0 (0 0) (1 1 2)) (1 (2 3) (3 4 5))) (1 (2 (4 6 7) (5 8)) (3 (6 9 10) (7 11)))) (1 (2 (4 (8 12 13)) (5 (9 14))) (3 (6 (10 15) (11 16 17)) (7 (12 18 19) (13 20 21))))) (1 (2 (4 (8 (14 22) (15 23 24)) (9 (16 25 26) (17 27))) (5 (10 (18 28) (19 29 30)) (11 (20 31)))) (3 (6 (12 (21 32) (22 33)))))) NP (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12 13) (7 14 15)))) (1 (2 (4 (8 16 17) (9 18 19)) (5 (10 20 21) (11 22 23))) (3 (6 (12 24) (13 25)) (7 (14 26 27) (15 28 29))))) (1 (2 (4 (8 (16 30 31) (17 32 33)) (9 (18 34 35) (19 36 37))) (5 (10 (20 38 39) (21 40)) (11 (22 41 42) (23 43 44)))) (3 (6 (12 (24 45 46) (25 47 48)) (13 (26 49 50) (27 51 52))) (7 (14 (28 53 54) (29 55 56)) (15 (30 57 58) (31 59 60)))))) @NP (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12 13) (7 14)))) (1 (2 (4 (8 15 16) (9 17 18)) (5 (10 19 20) (11 21))) (3 (6 (12 22 23) (13 24 25)) (7 (14 26 27) (15 28 29))))) (1 (2 (4 (8 (16 30 31) (17 32 33)) (9 (18 34 35) (19 36 37))) (5 (10 (20 38 39) (21 40 41)) (11 (22 42 43) (23 44 45)))) (3 (6 (12 (24 46 47) (25 48 49)) (13 (26 50 51) (27 52 53))) (7 (14 (28 54 55) (29 56 57)) (15 (30 58 59) (31 60)))))) DT (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1) (2 2))))) (1 (2 (2 (2 (3 3 4) (4 5 6)) (3 (5 7 8) (6 9))) (3 (4 (7 10) (8 11)) (5 (9 12) (10 13)))) (3 (4 (6 (11 14))) (5 (7 (12 15 16)) (8 (13 17) (14 18)))))) NNP (0 (0 (0 (0 (0 (0 0) (1 1 2)) (1 (2 3 4) (3 5))) (1 (2 (4 6 7) (5 8)) (3 (6 9 10) (7 11 12)))) (1 (2 (4 (8 13 14) (9 15 16)) (5 (10 17 18) (11 19 20))) (3 (6 (12 21 22) (13 23 24)) (7 (14 25) (15 26))))) (1 (2 (4 (8 (16 27 28) (17 29 30)) (9 (18 31 32) (19 33 34))) (5 (10 (20 35 36) (21 37 38)) (11 (22 39 40) (23 41)))) (3 (6 (12 (24 42 43) (25 44 45)) (13 (26 46 47) (27 48 49))) (7 (14 (28 50 51) (29 52 53)) (15 (30 54 55) (31 56)))))) CD (0 (0 (0 (0 (0 (0 0) (1 1)) (1 (2 2 3) (3 4))) (1 (2 (4 5 6) (5 7)) (3 (6 8) (7 9 10)))) (1 (2 (4 (8 11) (9 12 13)) (5 (10 14 15) (11 16))) (3 (6 (12 17 18) (13 19 20)) (7 (14 21 22) (15 23))))) (1 (2 (4 (8 (16 24)))) (3 (5 (9 (17 25)) (10 (18 26)))))) NN (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12 13) (7 14 15)))) (1 (2 (4 (8 16))) (3 (5 (9 17 18) (10 19 20)) (6 (11 21 22) (12 23))))) (1 (2 (4 (7 (13 24 25) (14 26 27)) (8 (15 28 29) (16 30 31))) (5 (9 (17 32 33) (18 34 35)) (10 (19 36 37) (20 38 39)))) (3 (6 (11 (21 40 41) (22 42 43)) (12 (23 44 45) (24 46 47))) (7 (13 (25 48 49) (26 50 51)) (14 (27 52 53) (28 54 55)))))) `` (0 (0 (0 (0 (0 (0 0)))))) '' (0 (0 (0 (0 (0 (0 0)))))) POS (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1)))))) PRN (0 (0 (0 (0 (0 (0 0) (1 1 2))))) (1 (1 (1 (1 (2 3))) (2 (2 (3 4) (4 5)) (3 (5 6)))))) @PRN (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1) (2 2)) (2 (3 3))))) (1 (2 (2 (3 (4 4 5)))) (3 (3 (4 (5 6))) (4 (5 (6 7)))))) -LRB- (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)))))) VBN (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1) (2 2 3)) (2 (3 4 5) (4 6)))) (1 (2 (3 (5 7)) (4 (6 8))) (3 (5 (7 9))))) (1 (2 (4 (6 (8 10) (9 11)) (7 (10 12 13) (11 14))) (5 (8 (12 15) (13 16 17)) (9 (14 18) (15 19 20)))) (3 (6 (10 (16 21 22) (17 23)) (11 (18 24) (19 25))) (7 (12 (20 26) (21 27)) (13 (22 28) (23 29 30)))))) NNS (0 (0 (0 (0 (0 (0 0) (1 1 2)) (1 (2 3 4) (3 5 6))) (1 (2 (4 7 8)) (3 (5 9)))) (1 (2 (4 (6 10 11) (7 12 13)) (5 (8 14 15) (9 16 17))) (3 (6 (10 18) (11 19 20)) (7 (12 21 22))))) (1 (2 (4 (8 (13 23 24) (14 25 26)) (9 (15 27 28) (16 29 30))) (5 (10 (17 31 32) (18 33 34)) (11 (19 35) (20 36 37)))) (3 (6 (12 (21 38 39) (22 40 41)) (13 (23 42 43) (24 44 45))) (7 (14 (25 46 47) (26 48 49)) (15 (27 50 51) (28 52 53)))))) VP (0 (0 (0 (0 (0 (0 0 1)) (1 (1 2 3))) (1 (2 (2 4 5) (3 6)) (3 (4 7)))) (1 (2 (4 (5 8 9)) (5 (6 10))) (3 (6 (7 11)) (7 (8 12 13) (9 14 15))))) (1 (2 (4 (8 (10 16 17) (11 18 19)) (9 (12 20 21))) (5 (10 (13 22)) (11 (14 23 24) (15 25 26)))) (3 (6 (12 (16 27 28) (17 29)) (13 (18 30))) (7 (14 (19 31 32) (20 33 34)) (15 (21 35 36) (22 37 38)))))) @VP (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12 13) (7 14 15)))) (1 (2 (4 (8 16 17) (9 18 19)) (5 (10 20 21) (11 22 23))) (3 (6 (12 24 25) (13 26 27)) (7 (14 28 29) (15 30 31))))) (1 (2 (4 (8 (16 32 33) (17 34 35)) (9 (18 36) (19 37 38))) (5 (10 (20 39 40)) (11 (21 41)))) (3 (6 (12 (22 42))) (7 (13 (23 43)) (14 (24 44 45) (25 46 47)))))) VBP (0 (0 (0 (0 (0 (0 0)) (1 (1 1))) (1 (2 (2 2)) (3 (3 3) (4 4)))) (1 (2 (4 (5 5) (6 6)) (5 (7 7 8))) (3 (6 (8 9)) (7 (9 10 11))))) (1 (2 (4 (8 (10 12 13) (11 14))) (5 (9 (12 15 16)))) (3 (6 (10 (13 17))) (7 (11 (14 18)))))) , (0 (0 (0 (0 (0 (0 0)))))) CC (0 (0 (0 (0 (0 (0 0 1)))) (1 (1 (1 (1 2))))) (1 (2 (2 (2 (2 3))) (3 (3 (3 4)))) (3 (4 (4 (4 5))) (5 (5 (5 6)))))) -RRB- (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1)))))) VBD (0 (0 (0 (0 (0 (0 0) (1 1 2)))) (1 (1 (1 (2 3))))) (1 (2 (2 (2 (3 4 5) (4 6 7)) (3 (5 8 9) (6 10 11))) (3 (4 (7 12 13) (8 14 15)) (5 (9 16 17) (10 18 19)))) (3 (4 (6 (11 20))) (5 (7 (12 21) (13 22 23)) (8 (14 24)))))) ADVP (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4) (3 5 6))) (1 (2 (4 7))))) (1 (1 (2 (3 (5 8) (6 9)) (4 (7 10) (8 11))) (3 (5 (9 12 13)) (6 (10 14) (11 15)))) (2 (4 (7 (12 16) (13 17)) (8 (14 18 19))) (5 (9 (15 20 21) (16 22 23)) (10 (17 24 25) (18 26 27)))))) RB (0 (0 (0 (0 (0 (0 0 1) (1 2)) (1 (2 3 4) (3 5 6))) (1 (2 (4 7 8) (5 9 10)) (3 (6 11 12) (7 13 14)))) (1 (2 (4 (8 15))) (3 (5 (9 16 17) (10 18)) (6 (11 19 20) (12 21))))) (1 (2 (4 (7 (13 22)) (8 (14 23))) (5 (9 (15 24 25) (16 26)) (10 (17 27 28)))) (3 (6 (11 (18 29) (19 30 31)) (12 (20 32 33) (21 34 35))) (7 (13 (22 36 37) (23 38 39)) (14 (24 40 41) (25 42 43)))))) TO (0 (0 (0 (0 (0 (0 0)))))) . (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1)))) (2 (2 (2 (2 2)))))) VBZ (0 (0 (0 (0 (0 (0 0)) (1 (1 1))) (1 (2 (2 2)) (3 (3 3 4)))) (1 (2 (4 (4 5)) (5 (5 6 7) (6 8))) (3 (6 (7 9) (8 10)) (7 (9 11 12))))) (1 (2 (4 (8 (10 13))) (5 (9 (11 14)))) (3 (6 (10 (12 15))) (7 (11 (13 16)))))) NNPS (0 (0 (0 (0 (0 (0 0) (1 1)) (1 (2 2))))) (1 (1 (1 (2 (3 3)) (3 (4 4 5)))))) SBAR (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1) (2 2)) (2 (3 3))) (2 (3 (4 4 5) (5 6 7)) (4 (6 8 9))))) (1 (2 (3 (5 (7 10) (8 11)))) (3 (4 (6 (9 12) (10 13 14))) (5 (7 (11 15)) (8 (12 16 17) (13 18)))))) PRP (0 (0 (0 (0 (0 (0 0 1) (1 2 3))) (1 (1 (2 4)))))) PRP$ (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1))) (2 (2 (2 2 3)))) (2 (3 (3 (3 4)))))) VB (0 (0 (0 (0 (0 (0 0) (1 1))) (1 (1 (2 2 3) (3 4)))) (1 (2 (2 (4 5))) (3 (3 (5 6)) (4 (6 7))))) (1 (2 (4 (5 (7 8 9) (8 10)) (6 (9 11 12) (10 13 14))) (5 (7 (11 15 16) (12 17)) (8 (13 18 19)))) (3 (6 (9 (14 20 21) (15 22 23)) (10 (16 24 25) (17 26))) (7 (11 (18 27 28) (19 29)) (12 (20 30 31) (21 32 33)))))) ADJP (0 (0 (0 (0 (0 (0 0 1))) (1 (1 (1 2)))) (1 (2 (2 (2 3) (3 4)) (3 (4 5 6))) (3 (4 (5 7) (6 8))))) (1 (2 (4 (5 (7 9) (8 10)) (6 (9 11) (10 12 13))) (5 (7 (11 14 15) (12 16 17)))) (3 (6 (8 (13 18 19) (14 20 21))) (7 (9 (15 22) (16 23 24)) (10 (17 25 26)))))) JJ (0 (0 (0 (0 (0 (0 0 1) (1 2 3)) (1 (2 4 5) (3 6 7))) (1 (2 (4 8 9) (5 10 11)) (3 (6 12) (7 13)))) (1 (2 (4 (8 14 15) (9 16 17)) (5 (10 18 19) (11 20 21))) (3 (6 (12 22 23) (13 24 25)) (7 (14 26 27) (15 28 29))))) (1 (2 (4 (8 (16 30 31) (17 32 33)) (9 (18 34 35))) (5 (10 (19 36 37) (20 38 39)) (11 (21 40 41) (22 42 43)))) (3 (6 (12 (23 44 45) (24 46 47)) (13 (25 48 49) (26 50 51))) (7 (14 (27 52 53) (28 54 55)) (15 (29 56 57) (30 58 59)))))) QP (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)) (2 (2 2))) (2 (3 (3 3))))) (1 (2 (3 (4 (4 4)) (5 (5 5 6))) (4 (6 (6 7)) (7 (7 8 9)))) (3 (5 (8 (8 10)) (9 (9 11)))))) @PP (0 (0 (0 (0 (0 (0 0 1))) (1 (1 (1 2 3)))) (1 (2 (2 (2 4))) (3 (3 (3 5))))) (1 (2 (4 (4 (4 6 7))) (5 (5 (5 8 9) (6 10)))) (3 (6 (6 (7 11)))))) MD (0 (0 (0 (0 (0 (0 0)) (1 (1 1))))) (1 (1 (1 (2 (2 2 3)))))) UCP (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)))))) @UCP (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1))))) (1 (1 (2 (2 (2 2)))))) VBG (0 (0 (0 (0 (0 (0 0) (1 1)) (1 (2 2) (3 3))) (1 (2 (4 4) (5 5)))) (1 (2 (3 (6 6 7))) (3 (4 (7 8)) (5 (8 9))))) (1 (2 (4 (6 (9 10)) (7 (10 11) (11 12))) (5 (8 (12 13) (13 14 15)) (9 (14 16 17) (15 18)))) (3 (6 (10 (16 19) (17 20))) (7 (11 (18 21) (19 22 23)) (12 (20 24 25) (21 26)))))) @SBAR (0 (0 (0 (0 (0 (0 0)) (1 (1 1))))) (1 (1 (1 (2 (2 2 3))) (2 (3 (3 4 5)))))) WHNP (0 (0 (0 (0 (0 (0 0)) (1 (1 1)))) (1 (1 (2 (2 2))) (2 (3 (3 3)) (4 (4 4)))))) @ADVP (0 (0 (0 (0 (0 (0 0) (1 1))))) (1 (1 (1 (1 (2 2)) (2 (3 3)))))) RBR (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1)))) (2 (2 (2 (2 2)))))) : (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1))) (2 (2 (2 2))))) (1 (2 (3 (3 (3 3)) (4 (4 4)))))) SINV (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1)))))) @SINV (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1)))) (1 (2 (2 (2 2))) (3 (3 (3 3))))) (1 (2 (4 (4 (4 4) (5 5)))) (3 (5 (5 (6 6 7))) (6 (6 (7 8) (8 9)))))) WP (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1)))))) WDT (0 (0 (0 (0 (0 (0 0 1))) (1 (1 (1 2)))))) JJR (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1) (2 2)) (2 (3 3) (4 4)))))) PDT (0 (0 (0 (0 (0 (0 0) (1 1)))))) RBS (0 (0 (0 (0 (0 (0 0)))))) @QP (0 (0 (0 (0 (0 (0 0)) (1 (1 1)))) (1 (1 (2 (2 2))) (2 (3 (3 3))))) (1 (2 (3 (4 (4 4))) (4 (5 (5 5)) (6 (6 6)))) (3 (5 (7 (7 7)) (8 (8 8) (9 9))) (6 (9 (10 10)) (10 (11 11 12)))))) @ADJP (0 (0 (0 (0 (0 (0 0)) (1 (1 1)))) (1 (1 (2 (2 2)) (3 (3 3))))) (1 (2 (2 (4 (4 4)) (5 (5 5)))) (3 (3 (6 (6 6)) (7 (7 7) (8 8))) (4 (8 (9 9)))))) FRAG (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1)))))) NAC (0 (0 (0 (0 (0 (0 0 1)))))) @NAC (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)) (2 (2 2)))))) WHADVP (0 (0 (0 (0 (0 (0 0))) (1 (1 (1 1)))))) WRB (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)))))) JJS (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1) (2 2))))) (1 (2 (2 (2 (3 3) (4 4)))))) $ (0 (0 (0 (0 (0 (0 0)))))) PRT (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)))))) RP (0 (0 (0 (0 (0 (0 0)))))) NX (0 (0 (0 (0 (0 (0 0))))) (1 (1 (1 (1 (1 1)) (2 (2 2)))))) @FRAG (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1 2)))))) WHPP (0 (0 (0 (0 (0 (0 0)))))) FW (0 (0 (0 (0 (0 (0 0)))))) SQ (0 (0 (0 (0 (0 (0 0)) (1 (1 1)))))) @SQ (0 (0 (0 (0 (0 (0 0)))) (1 (1 (1 (1 1)))))) @NX (0 (0 (0 (0 (0 (0 0)) (1 (1 1))))) (1 (1 (1 (2 (2 2)))))) SBARQ (0 (0 (0 (0 (0 (0 0)))))) @SBARQ (0 (0 (0 (0 (0 (0 0 1)))))) EX (0 (0 (0 (0 (0 (0 0)) (1 (1 1)))))) CONJP (0 (0 (0 (0 (0 (0 0)))))) WHADJP (0 (0 (0 (0 (0 (0 0)))))) SYM (0 (0 (0 (0 (0 (0 0)))))) # (0 (0 (0 (0 (0 (0 0)))))) @CONJP (0 (0 (0 (0 (0 (0 0)))))) LS (0 (0 (0 (0 (0 (0 0)))))) @WHADJP (0 (0 (0 (0 (0 (0 0)))))) INTJ (0 (0 (0 (0 (0 (0 0)))))) UH (0 (0 (0 (0 (0 (0 0)))))) WP$ (0 (0 (0 (0 (0 (0 0)))))) X (0 (0 (0 (0 (0 (0 0)))))) @WHNP (0 (0 (0 (0 (0 (0 0)))))) RRC (0 (0 (0 (0 (0 (0 0)))))) @WHADVP (0 (0 (0 (0 (0 (0 0)))))) LST (0 (0 (0 (0 (0 (0 0)))))) @X (0 (0 (0 (0 (0 (0 0)))))) @INTJ (0 (0 (0 (0 (0 (0 0)))))) @PRT (0 (0 (0 (0 (0 (0 0)))))) @LST (0 (0 (0 (0 (0 (0 0)))))) @RRC (0 (0 (0 (0 (0 (0 0))))))

Q: is it true that the infinity norm can be bounded using the $L_2$ norm the following way? Let $v \in \mathbb{R}^k$, and let $A \in \mathbb{R}^{m \times k}$ and let $B \in \mathbb{R}^{m \times n}$ such that each column of $B$, $B_i$, has $$||B_i||_2 \le 1.$$ Is it true that: $||v A^{\top} B||_{\infty} \le ||v A^{\top}||_2$ ? If the spectral norm of $A$ is such that $||A||_{\mathrm{spectral}} \le 1$, is it true that $||v A^{\top}||_2 \le ||v||_2$ ? Thanks. A: 1) Cauchy-Schwarz says $$|(v A^T B)_i| = |v A^T B_i| \le \|v A^T\|_2 \|B_i\|_2 \le \|v A^T\|_2$$ 2) Yes because the spectral norm is the operator norm corresponding to the $2$-norm on vectors, and $\|A\|_{\text{spectral}} = \|A^T\|_{\text{spectral}}$.

Overall guest ratings Review highlights Good to know Nearby About With a stay at Hanoi Charming Hotel, you'll be centrally located in Hanoi, steps from Vietnamese National Tuong Theatre and minutes from Hang Gai Street. This hotel is within close proximity of Hang Da Market and St. Joseph Cathedral. Rooms Make yourself at home in one of the 36 air-conditioned rooms featuring minibars and LCD televisions. Complimentary wireless Internet access keeps you connected, and cable programming is available for your entertainment. Private bathrooms with showers feature rainfall showerheads and complimentary toiletries. Conveniences include phones and desks, and housekeeping is provided daily. Amenities Make use of convenient amenities, which include complimentary wireless Internet access and tour/ticket assistance. Dining Take advantage of the hotel's 24-hour room service. Business, Other Amenities Featured amenities include a business center, dry cleaning/laundry services, and a 24-hour front desk. A train station pick-up service is provided at no charge.

Industrie 4.0_Glossary Everyone is talking about Industrie 4.0, digitization, big data and cloud technology just to name a few. Here you can find definitions and descriptions of most terms regarding the fourth industrial revolution. Robotics, automation technology and digitization will shape the next half-century in the same way that the Internet, smartphones and computers have shaped the past half-century. Our grandchildren will grow up as the first generation ‘R’ of ‘robotic natives’ – and we have to start taking responsibility for this today! Administration Shell _All data in a single shell. The administration shell is the virtual image of a hardware or software component in a production process, containing all the specific production data. This data opens the way to entirely new possibilities and added value in networked production. One decisive benefit is that all information – from CAD data and maintenance information through to configuration details – is located in one place without media discontinuity. Data and functions are available on the component itself, in the company network and/or in the cloud. This collection of information results in an all-encompassing knowledge which, once stored, can be made available for any user and any application. App _Expertise from the cloud. Apps for industrial scenarios – programs for defined, specific applications – are the answer that experienced specialists with extensive process and industry know-how offer up for customers’ challenges. Intelligent mobile apps make modern companies Industrie 4.0-capable. KUKA offers apps for a wide range of challenges across the entire value chain – in manufacturing, for example, their functionalities encompass Maintenance Services through Asset Management to Demand Planning. App Store _New capabilities on demand. Already equipped for the future: for generations, KUKA robot controllers have featured a modular and scalable configuration on the basis of mainstream technologies. Among other things, KUKA has thus created the foundation for making smart tools available via app stores and marketplaces. Intelligent digital forms or complete applications, such as those familiar today from well-known app stores for smartphones, tablets or computers, and which endow robots with new capabilities and functions on demand at the click of a mouse. For example, programs that only require entry of the desired parameters. With regard to Industrie 4.0, the immediate availability of new production capabilities will open up a whole new dimension of versatility for robots. Artificial Intelligence _Machines as intelligent partners. Artificial Intelligence (AI) is the step required for implementing the fourth stage of the robotic revolutions. It presupposes that machines, information systems and robots are capable of becoming still much more intelligent and responsive. In the fields of service robotics and home assisted living, these intelligent machines with their cognitive and sensitive capabilities will become increasingly important as helpers for humans. Today, these systems are still fully dependent on programming by humans. As the degree of autonomy of the systems increases, however, the issue of responsible management of artificial intelligence will become ever more pressing. Batch Size 1 _Unique, one-off products for everyone. Industrie 4.0 is creating the basis for implementing the highest levels of customization – all the way down to batch size 1 – within industrial manufacturing. This means high-quality, single-piece production at the price of current uniform, mass-produced goods. The networking of all systems involved in production, and their extreme flexibility, will make the fulfillment of individual customer requirements a matter of routine in the smart factory. While the desire for customized products is already a megatrend today, it will develop to become one of the decisive competitive factors in the near future. This trend not only offers new market opportunities for products, but also gives traditional industrial nations the option to return previously outsourced production capacity to locations in high-wage countries. Big Data _Data is the new oil. The term “Big Data” refers to quantities of data that are too large or too complex, that change too quickly or are too weakly structured for them to be evaluated with manual and conventional methods of data processing. In this context, experts talk about an inconceivably large data volume of currently more than eight zettabytes – with an increasing tendency. A substantial proportion of this already comes from the Internet of Things (IoT) and from the ever more numerous sensors in machines and vehicles. Data is increasingly being generated in real time. In connection with Industrie 4.0, however, it is the ability to evaluate and process this flood of data that is of paramount interest. That is how Big Data becomes Smart Data. The challenge is therefore not only for IT systems to be able to handle heterogeneous data correctly but also for them to analyze the data in order to create a reliable basis for business decisions – preferably in real time. Only in this way can processes be controlled intelligently and adapted to changing parameters. Taking the metaphor further, Big Data is thus the new oil of the 21st century. Cloud Robotics _Shared intelligence. Nowadays smartphones, tablets and computers utilize data and processing power from the cloud as a matter of course. In the context of Industrie 4.0, robots too will be able to access decentralized data in networks or in the cloud, thereby significantly boosting their performance and flexibility. The robot itself will only require a small chip to control functionality, motion and mobility. For the task at hand, specific services will be retrieved from the cloud or individual robots networked on an ad hoc basis to form temporary production teams. In this way, specialists will become universalists that can be used for a wide range of different manufacturing processes. Cloud robotics enables the implementation of a broad spectrum of different industry-specific applications via “Robotics as a Service®”. Another effect of the cloud: robots learn from one another. If one robot encounters an obstacle, for example, it posts this information to the connected systems, which can use it to respond intelligently to the obstacle. Collaborative Robots _Hand in hand. Collaborative robots – sometimes also known as “cobots” for short – are robots that are capable of human-robot collaboration (HRC) and work hand-in-hand with their human colleagues. As collaborative robots operate without physical safeguards, they have to permanently calculate the risk of colliding with humans, constantly checking this via the robot controller. The strict safety requirements have been redefined in the revised EN ISO 10218 standard, parts 1 and 2, and in the ISO/TS 15066 specification initially drafted in 2010. Besides the robot itself, the standard also covers the adapted end-of-arm tooling with which the robot performs its tasks, and the objects moved with it. With the LBR iiwa, KUKA has made the world’s first series-produced, collaborative lightweight robot for industrial applications ready for the market, thereby proving that the visions of Industrie 4.0 can be turned into reality. Committees _Strong alliances with KUKA. As a leading supplier of intelligent automation solutions, KUKA is directly involved in Industrie 4.0 and sees itself as responsible for forming strong alliances with the goal of actively shaping the factory of the future together with other key players. That is why KUKA is a member of all major national and international advocacy groups and committees, such as the Industrie 4.0 Platform coordinated by the German Federal Ministry for Economic Affairs and Energy, the Industrial Internet Consortium IIC, the OPC Foundation, the German industry association VDMA and the associations BITKOM and VDE. As a trailblazer for Industrie 4.0, KUKA is a sought-after interlocutor for decision-makers in the worlds of politics, research, and business. CPS (Cyper- Physical System) _Physical world meets virtual world. A cyber-physical system (CPS) is a “thing” in the Internet of Things (IoT). It is a combination of mechanical and electronic software components that communicate via a data infrastructure such as the Internet, react flexibly to external influences and exchange data with information systems and other CPSs. In future manufacturing facilities, cyber-physical systems will communicate with intelligent, networked industrial production and logistics units – also known as cyber-physical production systems (CPPS). The CPSs exchange information, trigger actions in production and reciprocally control themselves autonomously. This enables industrial processes in manufacturing, engineering, use of materials, supply chain management and life cycle management to be fundamentally restructured and optimized. Data Ownership _Who owns the data? The data must belong to their originator. A principle that is regrettably contested in the cloud. The open exchange of data and information, however, is a vital ingredient of Industrie 4.0. Information that was generated in the company prior to storage in the cloud is generally subject to local copyright laws. But if the data are created in the cloud, things become somewhat more ambiguous. Cloud providers have a different approach to dealing with user data in terms of access rights and at times create confusion as to ownership. It is therefore important to ensure transparency in the use of cloud services and to select secure encryption for all data sent to the cloud. This gives users control over their data and thus some form of ownership, regardless of legal issues. Particularly with a view to the horizontal networking of various companies within a production process, the question of data sovereignty is of central importance. With cloud solutions meeting the highest data security standards, KUKA offers unique platforms on the basis of which customers can exchange their own data with others or enrich them with new intelligence and additional information. Decentralized Intelligence _Intelligence evolves in the swarm. Decentralized intelligence will play an important role in Industrie 4.0: all parties can communicate with one another – workpiece with machine, machine with machine or with higher-level processes. No central “brain” will control and monitor the things, but rather autonomous production units will carry out this function for both heterogeneous and homogeneous teams. Decentrality makes for greater flexibility and quicker decisions. Intelligence evolves in the swarm or through joint networking with the cloud. Demographic Change _Society is getting ever older. In principle, the term “demographic change” is a neutral reference to any alteration in the age structure of a society. At present, however, it is being widely used as a synonym for increasing overaging in the industrial nations. A trend that is diametrically opposed to the rapid growth of the global population. By the year 2020, more than half of the German population will already be over 50 years of age. An ever smaller number of people in employment will have to generate the entire productive output for the social systems. This challenge can only be met if the remaining workers become considerably more productive than all generations before them. At the same time, opportunities must be created for older, experienced employees to participate in the world of work for longer. In order to make new working environments both highly productive and ergonomically beneficial for the labor force, KUKA is developing central key technologies for Industrie 4.0: collaborative robots, mobile assistance systems, autonomously controlled vehicles and smart, digitized automation solutions that support humans in the work setting, easing the workload in a variety of ways. Development Community _Innovating together. Intelligent solutions benefit from the expertise of many creative professionals. KUKA offers interested developers access to a powerful platform that serves as a central point of contact for exchanging a vast variety of information. This community assists its members with expertise, support and resources. This platform by developers for developers, a “KDN – KUKA Developer Network” so to speak, facilitates the very exchange of information that makes the added value of team creativity fruitful. The bundled expertise further strengthens the potential of the KUKA ecosystem. As an example, the platform makes it easier for start-ups to generate business processes and develop new business models. Digital Business Transformation _Creating the business of the future. Digitization is upending a large number of existing business models. Networked automation, intelligent processes, shared expertise across companies and industries, and a holistic view of the value chain with involvement of the ustomer – these are the characteristics of digitized business models. They open up completely new business opportunities and options, and offer competitive advantages – and only they make it possible to fulfill customer requirements optimally. There is no economically successful future without the digitization of business models. With its own Consulting center of excellence, KUKA is assisting companies along the way to a digital future. Digital Customer Journey _Journey through the digital business landscape. In general terms, the “Customer Journey” represents the sequence of (digital) contacts that a customer may encounter when dealing with another company. Depending upon the issue, the first point of contact and subsequent contacts may vary. If a purchase is intended, for example, the customer’s journey may start off with the links in a search engine request or a rating portal, followed by a shop visit. The app store and support website are other possible starting points that are implemented in industry. For business-to-business processes, the initial contact may take place through the business cloud instead of a website, for instance by visiting a developer platform. KUKA serves its customers as a hub for organizing their digital journey. Digital Shadow _Virtual image of real things. The digital shadow is a digital image of a real object. This data contains both the current status and the desired status of the object, the possible ways and processes for achieving this desired status, and the history of what the object has already gone through. Only through the combination of a digital shadow and a physical object does a smart thing result. Every physical product can be manufactured more efficiently and with higher quality in the digitized production facility if a digital shadow has been created for it and it bears its own specific DNA. Digital Value Chain _Transcending all boundaries. The digital supply chain merges the major business processes of all parties involved – from the suppliers to the manufacturer and the end customer. The potential of a digitized value chain lies primarily in the acceleration of the production and logistics processes, the reduction of effort for data acquisition and the optimization of data security and consistency. With integrated networking, the digital value chain is able to overcome current media discontinuity. One example from the field of procurement: where a steel-processing company previously had to activate a complicated process via different media for purchasing and replenishment, in the future purchasing will be automated on the basis of predefined parameters. Companies today are already making use of digital value chains to optimize individual production islands and processes within their organization. In the factory of tomorrow, the digital supply chain will also encompass global procedures across company boundaries, controlling them largely autonomously. As the most flexible machine ever conceived by man, the robot plays a central role in the digital supply chain. In its function as the core component of intelligent automation solutions, it increases the entrepreneurial freedom of action, creates competitive advantages, speeds up production processes and assures quality in the long term. Integration into the KUKA Connect platform accelerates the conversion to a digitized value chain. Digitization _Potential of the digital transformation. Converting real products and analog sequences into digital data and processes is referred to as digitization. In Industrie 4.0, people, machines and industrial processes are networked on the basis of cyber-physical systems incorporating state-of-the-art information and communications technology. In this context, the intelligent exchange and interpretation of data determine the entire life cycle of a product: from the idea to development, manufacturing, use and maintenance through to recycling. Production and logistics processes will be globally networked beyond the factory gates in the future for the purpose of optimizing the flow of materials, detecting non-conforming parameters at an early stage and enabling a highly flexible reaction to changing customer requirements and market conditions. Edge Cloud Gateway _The Cable Modem of the shop floor. Much like a DSL Router connects peripherals to the Internet, an edge cloud gateway connects the various devices at the production level to a network or a cloud environment via EtherNet or wireless / mobile connection, so that data can be collected, exchanged and processed. Features such as secure communication and quality of service are implemented. Flexibility _Flexibility in all dimensions. Flexibility is the ability to react quickly to changing influences. In the smart factory, utmost flexibility results primarily from the combination of IT technologies, such as the Cloud and Big Data, with intelligent, generic production units incorporating robots and autonomously controlled mobile units. The factory of the future will not have any predefined routes or rigid processes. Mobile units will equip robots “on the fly” with other tools, enabling them quickly to carry out new tasks or process other workpieces. The smart factory is therefore able to manufacture different products or product versions without any significant retooling times. It thus completely redefines the concept of flexibility in production. Framework _The programming kit. Generally speaking, a framework is a defined structure for the programming of software or dynamic websites. So-called “pre-programs” allow developers to save time: frequently recurring program steps are already pre-programmed and can simply be called up when needed. Thus, it is no longer necessary to repeat certain functions and computer tasks. Frameworks provide valuable support particularly for quickly programming apps. In robotics, a framework provides software components for industrial use. This includes algorithms, libraries and methodological know-how. Standardized interfaces and appropriate middleware make it easy to replace hardware and software components. This simplifies and speeds up application development in automation. Home Assisted Living _Living independently – even in later life. Our society is becoming ever older: in 2035, one in every three Germans will be over 60 years of age. Everybody wants to stay autonomous and active for as long as possible when old, however. Home assisted living enables the elderly to continue an independent life within their own four walls. Besides service robots, which take care of household activities, and smart home applications, services in the medical and nursing sectors will be a part of daily life in the future. With the aid of intelligent robots, rehabilitation treatment, for example, will be possible at home. Mobility assistants will help people to remain agile into old age, improving the quality of millions of lives. Horizontal Integration _Dynamic company networks. Exact coordination is not only indispensable for internal process optimization within a company, but also between all companies involved in the value chain. This horizontal integration – networking between different enterprises and their assets – is the starting point for the flexible design of their shared value creation processes. In the era of Industrie 4.0, companies will form dynamic networks in the future, linking order-specific and product-specific capacities in virtual production communities. Current data from the production-relevant processes will enable fast and precise reactions – for instance to planning changes or unexpected events occurring inside or outside an individual company. Production and logistics processes adapt to the real situation in real time, boosting the long-term flexibility and efficiency of the companies acting within an integrated concept. Process analyses to increase productivity involve many different components – and thus also their manufacturers. See also Manufacturing Ecosystem. Individualized Production _Meeting every customer requirement. Individualized, or customized, production refers to the concept of an intelligent, highly automated production system that allows high variance and dynamism in the product range with production costs at the level of mass production. The goal is to resolve the conflict between the customer’s desire for individualization and the process efficiency of production in an industrial setting. A batch size of 1 is the highest level of customized production. Besides proprietary solutions in the automotive sector, Industrie 4.0 with its universally networked production environments represents the world’s most advanced approach for implementing customized production. Industrie 4.0 _Production meets digitization. Industrie 4.0, Smart Production or Internet of Things (IoT) – even if the names and terms used vary from one country to another, they all share the same goal. What is called for here is nothing less than a long-term transformation of our global perception of industrial production through the seamless connection of the digital and real worlds. KUKA is at the interface between these two worlds and is playing a decisive role in advancing this transformation as a thought leader and trailblazer for Industrie 4.0. It was back in the 1990s that KUKA as a first mover recognized the potential to be gained by combining the world of IT with conventional automation technologies. The company was also the world’s first robot manufacturer to develop open, interoperable and flexible systems on the basis of standardized mainstream technologies and to make them ready for the market. In collaboration with experts from diverse sectors, KUKA is now already implementing highly flexible, digitized manufacturing processes that will open up new opportunities in a competitive environment and lastingly change the way we work and produce. KUKA supports companies with the digitization of production across industries – from small and medium-sized companies to large OEMs. This provides a platform for anyone who is ready to step into the world of Industrie 4.0. Internet of Automation (IoA)/ Internet of Robotics (IoR) _Basis for efficient production. Both the Internet of Automation (IoA) and the Internet of Robotics (IoR) make use of defined open communications and data standards to network interoperable production processes even across company boundaries. In the IoR for example, KUKA robots, the KUKA App Store, middleware and monitoring tools are networked to form a highly efficient production environment in which analog and digital devices can easily communicate with one another. In the near future, it will be possible for all the cyberphysical elements involved in the automated manufacturing process to be networked in the IoA and to communicate with the IoR. Extensive standardization of protocols and technologies is required for this Internet-based infrastructure. See also Committees. Internet of Things (IoT) _Everything communicates with everything else. Like Industrie 4.0, the Internet of Things (IoT) presupposes a network of physical objects – devices, vehicles, buildings and other items – which are fitted with electronic components, software and sensors, all of them being linked interoperably via the Internet. Unlike Industrie 4.0, the IoT rather non-selectively refers to all things that could be connected to the cloud. The IoT thus also encompasses the private domain, including, for instance, the already well-known “smart home” applications. Strictly speaking, the smart factories of Industrie 4.0 along with all their production and logistics processes are a part of the IoT. Experts forecast that the IoT will comprise 50 billion objects by the year 2020. Interoperability _Everything works together. Interoperability (IOP) describes the ability of an object, device or machine to communicate with other things in the network. It must be able to do so regardless of whether the devices are from the same or different manufacturers. Interoperability is a fundamental precondition for creating a layer that enables cyber-physical systems to be interconnected such that interactions are possible without the participants knowing which technologies the implemented devices are based on. It is also the basis for the capability of the things in the network to communicate without any restrictions and to act intelligently as a swarm. See also Committees. KMP omniMove _Strong on the move. Wherever size, load-bearing capacity and precision are required, the KMP omniMove heavy-duty mobile platform is in its element. Individually or as part of a fleet, the KMP omniMove can effortlessly move beneath, and lift, workpieces weighing more than 90 tonnes and move in any desired direction with millimeter precision from a standing start, thanks to the multidirectional omniWheels. The KMP omniMove can move autonomously, under guidance along a programmed path or conventionally under manual control. KMR iiwa _New horizons. Shorter response times and greater flexibility going beyond full automation: these are the requirements of markets that are changing at an ever faster pace. The industrial manufacturing of the future will require modular, versatile and, above all, mobile production and manufacturing concepts. That is why the KMR iiwa unites the sensitive and compliant LBR iiwa lightweight robot with KUKA’s mobile platform technology to form a new, intelligent and fully mobile combination that can work in the vicinity of humans. Just like humans, the KUKA Mobile Robots (KMR) can also track moving workpieces, move around them freely and link solitary production islands to form new production units. KMR QUANTEC _Large-dimension e-mobility. Anyone thinking big and looking for flexible mobility will find the perfect powerful partner in the KMR QUANTEC. The combination of KUKA six-axis robots, mobile platforms, high-performance energy storage units and industrialgrade components creates a mobile solution for virtually any scenario. Despite its strength, the KMR QUANTEC is characterized by outstanding precision and maximum electromobility. Its small power plant supplies it with electricity for a full eight-hour shift. The position and number of robots installed are variable, as too are the size and payload capacity of the platform. Grippers, power-hungry tools and special equipment can also be transported on the KMR QUANTEC and continuously supplied with power. KUKA flexFELLOW _Mobility on demand. The KUKA flexFELLOW automation unit can be moved manually and allows ad hoc localized variation of the degree of automation in production. Without the need to alter the production layout, for example, the KUKA flexFELLOW is able to open the safety doors at machines, which it can then load and unload independently. It can also assist the operator in performing work steps in ergonomically unfavorable situations. In alternating operation, humans and robots can ideally complement one another. The combinability of manual and automatic tasks means that production can be optimally adapted to the specific requirements. KUKA LBR iiwa _Robotic colleague. KUKA is starting a new chapter in the history of industrial robotics with the lightweight robot LBR iiwa (intelligent industrial work assistant). As the first series-produced sensitive robot for human-robot collaboration (HRC), the LBR iiwa is tapping new applications that were previously closed to automation. Thanks to its sensory capabilities, it can intuitively learn new tasks on an ad hoc basis, simply through being guided by its human partner. The machine becomes a “robotic colleague”. It works hand in hand with the operator, thereby enabling him to work more efficiently, more ergonomically, more precisely and with greater concentration. As a robot that can genuinely be deployed universally, it is defining new standards on the road to the fourth industrial revolution. Logistics _Now. Everything. Always. Customized products and same-day delivery – customers have a growing expectation that everything will be available in all places, at all times. This ubiquity places the utmost demands on the logistics and process chains and is increasingly embracing the stationary retail sector and the structure of merchandise flows. The boundaries between individual delivery channels are successively vanishing and modern distribution centers are often being set up directly in metropolitan areas thanks to the reduced space requirements. Changes that can only be addressed through highly transparent, digitized networking of production and logistics. In this context, KUKA sees itself as a solution provider translating the individual requirements of the market participants into flexible, networked and software-supported logistics concepts. M2M Communication Protocol _The universal communicator. Machine-to-machine communication, or M2M, is the automatic exchange of information between machines. These may be vastly different end devices, from production machines through vending machines to vehicles or household appliances. A standardized set of rules is needed for M2M communication to work – in modern terms: a protocol. Such a standardized software interface enables universal communication. The most promising protocol at this time is the Open Platform Communications Unified Architecture (OPC UA). Machine Learning _Knowledge through experience. Intelligent machines garner their knowledge through experience. In the case of networked machines, it is irrelevant whether the experience is their own or originates from swarm intelligence. An artificial system always learns by comparing the desired objective and any anomalies that occur. It can recognize correlations, patterns and general rules, draw conclusions from them and modify its future behavior, this synthetic process being referred to as machine learning. Especially in unstructured environments and with highly flexible processes like Industrie 4.0, machine learning in a swarm or in the cloud is an effective method of adapting production processes intelligently and autonomously to the individual framework practically in real time. Manufacturing as a Service (MaaS)/ Robotics as a Service (RaaS)® _Access rather than ownership. Digitization has substantially changed the approach to physical possession. This is increasingly being replaced by temporary access to goods or services. The best example: music streaming. What has already become an everyday situation in many consumer segments will also revolutionize the industrial environment over the next few years. As the name implies, manufacturing processes are provided as a service when offering “Manufacturing as a Service”: the machine does not change ownership, and only the service of the machine is paid for (for example in the form of a cost-per-uptime model). What applies to complete production systems will, in the future, also hold true for individual elements within a manufacturing facility – for robots, for example. On the basis of a “pay-per-use” model, it will not be the physical object itself that is purchased in conjunction with “Robotics as a Service®”, but rather its performance, such as weld spots in vehicle body production, for instance. The smart factory of the future integrates these services seamlessly into its production processes and thus has the capability of reacting to varying capacity requirements and goods flows exceedingly flexibly and efficiently while conserving resources at the same time. Ultimately, these business models uproot traditional ideas of what a value chain may look like: the principle of shared production will become more important, allowing the end user to act as a co-producer. If desired, KUKA Consulting analyzes the business models of companies to determine whether they can be marketed profitably as a service (“Product as a Service”). Manufacturing Ecosystem _Expertise is the raw material for creating value efficiently. If the best available expertise can be focused upon every single process step and every single level of activity, all parties involved benefit: product designers, developers, manufacturers, logistics specialists and end customers. A complex network that optimally brings together all relevant expertise in terms of personnel, economy and technology is considered an ecosystem. This comprises hardware and software resources, process and project know-how, consulting expertise and complex networked architectures of production and information technology. What is important for such an interdisciplinary system is that the combined expertise is not just limited to a single company, but that the best possible personnel and the best available knowledge on the market is brought together under one project umbrella. As an experienced player across industries, products and processes at the interface of OT and IT, KUKA is in a unique position to “fuse in” the ideal experts and expertise for any given project. This yields an unbeatable team of resources provided by the customer, KUKA and external partners, with superior know-how and advanced technology from the shop floor to the business model level. Ecosystems add value beyond just optimizing the value creation processes: this bundled creativity allows for new products, solutions and business models to be developed jointly. Megatrends _What makes the world go round. Customization, digitization, responsible use of natural resources and demographic changes are the megatrends that will need to be mastered in the coming decades. With a forecast world population of eight billion by the year 2025 and ten billion by 2060, ever more customer requirements of increasing diversity will need to be satisfied. At the same time, demographic changes will be confronting industrialized and emerging countries with economic and social challenges over the long term. Humanity is thus facing a fundamental paradigm shift which will undoubtedly have far-reaching consequences for our worldwide economic systems. That is why Industrie 4.0 does not describe a purely technical innovation scenario but rather a way in which intelligent technology can help to overcome the global challenges of the 21st century. As a thought leader and trailblazer for Industrie 4.0, KUKA is working on production environments that increase economic efficiency while also using resources responsibly, which make high-quality goods more affordable and which are instrumental in sustainably improving human working conditions in factories. Mobility _Robots on the advance. The production of the future requires greater flexibility – for both faster manufacturing and customized products. One indispensable prerequisite for creating a higher degree of flexibility in industrial environments is greater mobility. This can be achieved by means of episodic, periodic or permanent mobility concepts that each bring about acceleration in industrial production in their own specific ways. For this reason, KUKA is paving the way for this key technology with platforms that can move intelligently and autonomously in order to transport things or machine workpieces. They can find their own way if necessary, aligning themselves with workpieces to within a millimeter or even moving the robot to them. They are also capable of responding autonomously and adaptively to changing conditions in complex logistics and production chains. Mobile automation solutions from KUKA show today what the future holds in store for highly flexible production in Industrie 4.0. Monitoring & Stream Analytics/ Micro Services _Data in real time. Monitoring and stream analytics compare and analyze data that is available to the smart factory from diverse sources – devices, sensors, infrastructure, etc. In real time they make comparisons with data records from the past, recognize anomalies and are able to categorize faults with the aid of machine learning. On this basis, intelligent systems initiate immediate countermeasures and generate forecasts and recommended actions for the future. On Premises _Software on board. On-premises solutions rely on the infrastructure of the company’s in-house IT. The deployment model of such software can be designed as a service and thus take on the function of a “local cloud”. People First _Production and products for people. In the factory of the future, the focus of thought and action will shift to the human worker with his changed requirements, desires and capabilities. Networked and flexible production technologies combine the cost advantages of mass production with the customization potential of craft manufacturing. This means that the customers’ desire for high-quality individualized products does not lead to higher prices, as is currently the case. But humans are at the center of the future world of goods not only in their role as customers. Intelligent robots that collaborate with humans, and mobile assistance systems will improve the world of work in many different ways. They handle heavy loads, carry out activities that are not ergonomic or are simply too dangerous for people and perform tasks which require levels of precision and speed of which humans are not capable or which are cognitively stressful. Predictive Maintenance _Eliminating static maintenance intervals. Dependable production planning and maximum machine availability through the avoidance of unscheduled downtime are the practical advantages of what is internationally known as predictive maintenance. On the basis of real-time data, all relevant parameters of the machines involved in the manufacturing process are acquired and evaluated for anomalies by means of stream analytics. In a subsequent machine learning process, specific fault patterns and the causes of a problem are detected in good time. This results in fewer rejects and maximum availability over the entire life cycle of the production line. The requirements on machine operating times vary depending on the specific branch of industry and its product cycles. While this could be up to 30 years in the aerospace sector, it is a matter of just a few months in the case of fast-moving goods, such as smartphones. With the aim of enabling an accurate assessment of the future performance of the machine or one of its components, intelligent predictive maintenance systems interconnect the largest possible amount of data from decentralized sources for the purpose of analysis. Preventive Maintenance _Efficiently controlling necessary downtime. Preventive maintenance is usually carried out on the basis of fixed contracts, which provide for an inspection of critical elements of the production systems at regular intervals or at fixed times. This may include analyses and cleaning the machines, for example. One of the advantages offered is the easy-to-plan and predictable maintenance times, which serve to prevent unforeseen downtime and ultimately save costs. Reshoring _Available is the new cheap. Whereas in the past purchase decisions were often primarily made on the basis of the lowest price, in the future it will be the product that is available most quickly and with a high level of customization that will be at the top of consumers’ shopping lists. This necessitates new manufacturing and marketing methods and structures that will only become possible as a result of networked production in smart factories. Short distances will be an important factor in achieving fast availability. Due to the high degree of automation, production steps that are currently outsourced to low-wage countries can be repatriated to high-wage countries in a process known as “reshoring”. Irrespective of wage structures, intelligent automation allows cost-efficient and high-quality production in the vicinity of the consumers. Resource Efficiency _Sustainable production. The ability of humanity to handle the future will be determined by a responsible and sustainable approach to natural resources. In a just world, it may be assumed that ever more people will want to be supplied with ever better products. Flexible, intelligent and networked production as envisaged in Industrie 4.0 offers the opportunity of using raw materials more efficiently and more sustainably along the entire value chain and recycling them to a great extent for the sake of the planet. Robofacturing _Craftsmanship meets robotics. Unlike a conventional factory, with its high degree of uniform mass production, craft manufacturing combines the virtues of skilled craftsmanship with a low level of mechanization. Craft-manufactured products convince with high-quality and a distinctly unique character. Robofacturing unites the advantages of craft manufacturing with the low price of a mass product, making individual and high-quality products affordable for large parts of the global population. Robotic Governance _Creating a responsible future world for generation “R”. Robotic governance is a concept which, among other things, considers the ethical/moral, socio-cultural, socio-political and socio-economic effects of robotics on society and provides a framework for solving problems resulting from these changes. The governance principles include accountability, responsibility, transparency of structures and fairness. In this way, robotic governance helps to create a sustainable and responsible future world for the upcoming generation “R”. Robotic Natives _Robots as natural companions. Future generations, “robotic natives”, will see robots as the state of the art, as a lifestyle, or quite simply as normal. Just as commonplace as smartphones and the Internet, for example, are for the digital natives of today. They will have overcome the old human versus machine antagonism. The robot-oriented generation will shape a society that not only works differently, but also thinks differently. They will see the capabilities of robots as universal, networkable services that can be requested via the Internet and flexibly adapted to the requirements and desires of the individual at the click of a mouse. While nowadays robots are primarily known as work assistants in industrial processes, in the future they will be found in all areas of our daily lives as driverless cars, robo-furniture, carebots and a wide range of home and personal assistants. By 2050, a robot in every household will be part of everyday life. Security _No success without security. The issue of safety and security is one of the most multifaceted business aspects of modern companies. In the separate worlds of production (OT) and information (IT), a basic distinction is made between safety and security: in OT contexts, the term “safety” is used to refer to the safeguarding of people and machinery as well as the availability and reliability of production equipment. In the IT environment, the term “security” primarily relates to data security, integrity and confidentiality. OT and IT are merged in the networked world of Industrie 4.0, which brings the issue of safety and security to a new level of complexity. The use of cloud models and the networking of entire companies and organizations exacerbates the security aspects (see also Data Ownership). KUKA Consulting offers well-founded advice on all safety and security questions and can draw on a broad network of expertise, including strategic alliances with international experts. Service Robotics _Robots enter daily life. Even today, useful robotic assistants are making everyday life easier. Small, specialized service robots, for example, have long since established themselves in our private sphere. They are deployed as assistants in the home – vacuuming, mowing the lawn or cleaning windows. As yet, their capabilities are mostly limited to a single task. However, they do demonstrate one thing: collaboration between humans and robots works in everyday life. Thanks to the progressive development of service robotics, robots will shape daily life in the future in various ways. Whether it be as a nursing robot in clinical settings, as helpers for the elderly at home or as assistants in many other areas which today sound futuristic. For the robotic natives of tomorrow, service robots will be as self-evident as smartphones are for people of the present. Smart Data _Intelligent data exchange. If Big Data is the oil of the future, then Smart Data is the fuel that drives the production of the future. Currently, data is just data. To turn it into information, it must be interpreted. This is the step from perception (recognizing) to cognition (understanding). Books, for example, are at first merely collections of letters. They only become knowledge when they are processed and interpreted in the brain. KUKA is developing smart data technologies for the digital domains in the age of Industrie 4.0. In the context of intelligent automation, the central focus is on the topics of data communication, process modeling, machine learning, autonomous self-configuration and process optimization. Smart Factory _Intelligent and self-organizing. The intelligent factory of the future is a production facility in which manufacturing systems, robots, logistics systems, products and their components are largely able to organize themselves autonomously. The smart factory is undergoing a paradigm shift towards an entirely new production logic: smart products, components, tools and machines are unambiguously identifiable, can be localized at all times and are aware of their history, their current status and multiple ways to the desired goal. With the smart factory’s high degree of flexibility, customization with a batch size of 1 will become reality in the context of industrial mass production. To achieve this, the production systems must, on the one hand, be networked vertically, for example with business processes within factories and companies. On the other hand, they must also be linked horizontally across company boundaries – from the purchase order through to outbound logistics – to create distributed value creation networks that can be controlled in real time. Smart Platforms _Intelligent and flexible. New, intelligent platforms will be created for the implementation of Industrie 4.0. They will support collaborative industrial processes and use their services and applications to network people, things and systems. The result will ensure greater flexibility and a continuous flow of information: smart platforms will document the entire business process, work safely and reliably at all levels, and support mobile end devices and collaborative production, service, analysis and forecast processes along the entire digital supply chain. For the smart factory, KUKA already has modular software architectures in its portfolio, based on mainstream technologies and prepared for the entire evolutionary process of Industrie 4.0. The Java platform of KUKA Sunrise makes it ideally suited to future app-based programs. Such a platform sets the stage for crossindustry digitization and thus increased value creation. This forms the foundation for the smart factory of the future. Social Machines _Interconnected. Intelligent. Flexible. Machines in production that are intelligently interconnected, communicate with one another and can instantaneously react to deviations and changes in an independent and situation-based manner are called social machines. They are part of the Industrie 4.0 vision. The underlying idea is that machines are able to share their knowledge like in social networks – information about themselves as well as experiences and “lessons learned” from their processes. At the same time, social machines coordinate the information received and learn from the network too. Similar to Facebook users, they independently obtain information from the Internet and connected social machine networks. Through swarm experience, they are aware of the best parameters for machining a particular material, for example, and they exchange them with “befriended” machines. Standardization _General requirements for reliable interaction. In the course of any technical evolution, different solutions, formats and approaches usually compete with one another – developed and propagated by various factions, committees or companies. Only standardization, with its exact and binding definition of framework parameters and the possible interfaces, makes it possible to create appropriate expansions, counterparts and communication bridges to a new technology. Of particular importance in this context with regard to the implementation of Industrie 4.0 are the new definitions of safety in the area of human-machine collaboration and the standardization of interoperability in the area of data exchange. KUKA is striving for OPC UA to be established as one of the future standards. This protocol not only transports machine data, parameters, process values and measured data, but can also, in combination with other standards, define them semantically in a way that is machine-readable. Standards _The basis of all constructive cooperation. Standards are the elementary basis for breaking down barriers in the globalized world of Industrie 4.0. As a world leader in automation, KUKA sees itself as responsible for playing a key role in shaping the areas of standards and standardization. To this end, KUKA is driving forward the harmonization of communications, data exchange and safety, for example, in the field of direct human-machine collaboration. In this way, it is creating new standards to ensure interoperability in the smart factory of tomorrow. As one of the key players in the paradigm shift from the third industrial revolution to the fourth, KUKA is thus laying the foundation for successful cooperation between all those involved. Time to Market _Meeting customer requirements more quickly. The time to market (TTM) denotes the length of time from development of a product to its availability on the market. In the factory of the future, this time, which is often decisive for the sales success of a product, will be significantly shortened. Positive effect: changing requirements and trends in increasingly volatile markets can be met with corresponding products much more quickly than previously. Traceability _Keeping track. Traceability here refers to the ability to fully trace all raw materials, producers, upstream suppliers, individual parts or assemblies as well as the complete product and its consumers in the digital value chain. It is possible at all times to determine when, where and by whom the goods were produced, processed, stored, transported, used or disposed of. Irrespective of whether an individual part or a finished product is concerned, a distinction is made between two directions of traceability: from the manufacturer to the consumer and from the consumer to the manufacturer. Value Data _From worthless data to valuable information. Individual data is like isolated objects in a void. They only make sense in context, and meaningful data can yield information that increases in value through being made exploitable for various purposes. Without being assigned, processed, compared, etc., data about the actual states of different parameters of a machine are nothing but loosely connected numbers. Intelligent evaluation of these data can provide valuable assets, for example by using forecasts that minimize or prevent maintenance and downtime. Vendor Cloud _Cloud with limited access. Large manufacturers with extensive supply networks avoid amalgamating different production data in the cloud, as they are concerned that this could reveal information about the production processes. But since the various suppliers and partners need important information from the production process, the manufacturers implement their own production cloud on the Internet or on their premises. In this vendor cloud, the production company can give suppliers access to exactly the data and information that are relevant to them, while all the rest remains hidden. KUKA can develop, implement and operate such cloud solutions for OEMs. KUKA is the ideal partner for operating vendor clouds thanks to its many years of experience with similar projects (including infrastructure work and coordination with suppliers and customers). Vertical Integration _Internal networking optimizes the production landscape. Networking the various assets within a company into an efficient production unit is one of the prerequisites for creating a smart factory. The different manufacturing stages with their respective specific assets can thus be merged into an integrated production process, for example. The networking of individual levels and assets via a cloud enables company-wide horizontal integration of the value chain as described in the Industrie 4.0 concept. For example, it allows integration into the KUKA ecosystem with all the resulting benefits (see Manufacturing Ecosystem).

{ "_from": "etag@~1.8.1", "_id": "etag@1.8.1", "_inBundle": false, "_integrity": "sha1-Qa4u62XvpiJorr/qg6x9eSmbCIc=", "_location": "/etag", "_phantomChildren": {}, "_requested": { "type": "range", "registry": true, "raw": "etag@~1.8.1", "name": "etag", "escapedName": "etag", "rawSpec": "~1.8.1", "saveSpec": null, "fetchSpec": "~1.8.1" }, "_requiredBy": [ "/send" ], "_resolved": "https://registry.npmjs.org/etag/-/etag-1.8.1.tgz", "_shasum": "41ae2eeb65efa62268aebfea83ac7d79299b0887", "_spec": "etag@~1.8.1", "_where": "/Users/meitu/hexo/hexo-theme-huweihuang/node_modules/send", "bugs": { "url": "https://github.com/jshttp/etag/issues" }, "bundleDependencies": false, "contributors": [ { "name": "Douglas Christopher Wilson", "email": "doug@somethingdoug.com" }, { "name": "David Björklund", "email": "david.bjorklund@gmail.com" } ], "deprecated": false, "description": "Create simple HTTP ETags", "devDependencies": { "beautify-benchmark": "0.2.4", "benchmark": "2.1.4", "eslint": "3.19.0", "eslint-config-standard": "10.2.1", "eslint-plugin-import": "2.7.0", "eslint-plugin-markdown": "1.0.0-beta.6", "eslint-plugin-node": "5.1.1", "eslint-plugin-promise": "3.5.0", "eslint-plugin-standard": "3.0.1", "istanbul": "0.4.5", "mocha": "1.21.5", "safe-buffer": "5.1.1", "seedrandom": "2.4.3" }, "engines": { "node": ">= 0.6" }, "files": [ "LICENSE", "HISTORY.md", "README.md", "index.js" ], "homepage": "https://github.com/jshttp/etag#readme", "keywords": [ "etag", "http", "res" ], "license": "MIT", "name": "etag", "repository": { "type": "git", "url": "git+https://github.com/jshttp/etag.git" }, "scripts": { "bench": "node benchmark/index.js", "lint": "eslint --plugin markdown --ext js,md .", "test": "mocha --reporter spec --bail --check-leaks test/", "test-cov": "istanbul cover node_modules/mocha/bin/_mocha -- --reporter dot --check-leaks test/", "test-travis": "istanbul cover node_modules/mocha/bin/_mocha --report lcovonly -- --reporter spec --check-leaks test/" }, "version": "1.8.1" }

Spin-polarized scanning tunneling spectroscopy of self-organized nanoscale Co islands on Au(111) surfaces. Magnetic monolayer and bilayer Co islands of only a few nanometer in size were grown by atomic deposition on atomically flat Au(111) films. The islands were studied in situ by scanning tunneling microscopy (STM) and spectroscopy at low temperatures. Spin-resolved tunneling spectroscopy, using an STM tip with a magnetic coating, revealed that the Co islands exhibit a net magnetization perpendicular to the substrate surface due to the presence of spin-polarized d-states. A random distribution of islands with either upward or downward pointing magnetization was observed, without any specific correlation of magnetization orientation with island size or island height.

Introduction ============ Catalytic transformations involving gold carbenes are arguably the most important aspect of homogeneous gold catalysis.^[@cit1]^ Recently, the possibility of forming an α-oxo gold carbenoid species *via* gold-catalyzed intramolecular or intermolecular alkyne oxidation by a N--O bond oxidant (initially a sulfoxide), pioneered by Toste and Zhang,^[@cit2a],[@cit2b]^ represents a significant advance in gold carbene chemistry, and various efficient synthetic methods have been developed based on this strategy.^[@cit2]^ Compared with intramolecular alkyne oxidation, the intermolecular approach offers much greater flexibility as no tethering of the oxidant is required, and therefore it is more synthetically useful.^[@cit3]^ However, this intermolecular approach is obviously not atom-economic as the reaction produces a stoichiometric amount of pyridines or quinolines, the reduced form of the corresponding pyridine *N*-oxides or quinoline *N*-oxides, as waste (eqn (1)),^[@cit4]^ which may even deactivate the gold catalyst *via* coordination.^[@cit5]^ Access to the related α-imino gold carbenes *via* gold-catalyzed nitrene transfer to alkynes, however, remains a highly challenging task. Here, it should be noted that: (1) the nitrene moiety is delivered *via* an outer sphere attack and no gold nitrene complex^[@cit6]^ is involved in this case; this mode of nitrene transfer is distinctively different from many well-established nitrene transfer reactions;^[@cit7]^ (2) this protocol would present alkynes as equivalents of α-diazo imines, which are difficult to access as α-diazo imines can readily cyclize into the corresponding 1,2,3-triazoles. To date, only limited success has been achieved in this type of gold-catalyzed nitrene transfer, mainly by the intramolecular reaction of alkyne and azide.^[@cit8]^ For example, Toste and co-workers used azide as an effective nitrene equivalent and realized the first protocol for the generation of α-imino gold carbenes in 2005.^[@cit8a]^ Later, elegant studies on the synthesis of indoles from alkynyl azides were demonstrated by Gagosz^[@cit8c]^ and Zhang,^[@cit8d]^ independently. Recently, several studies have invoked the intermolecular transfer of nitrene to alkynes by the use of iminopyridinium ylides as nitrene-transfer reagents, as disclosed by the groups of Zhang,^[@cit9]^ Davies,^[@cit9b],[@cit9c]^ and Liu.^[@cit9d]^ However, similar to those of the above-mentioned gold-catalyzed intermolecular alkyne oxidations, a stoichiometric amount of pyridine was produced as the waste in these cases. Therefore, the exploration of intermolecular approaches to the generation of α-imino gold carbenes, especially in an atom-economic way, is very attractive to researchers. We envisioned that the α-imino gold carbene intermediate **B** might be generated through the gold-catalyzed intermolecular reaction of ynamides **1** ^[@cit10]^ with isoxazoles **2**, which could be obtained in an efficient and modular manner following the synthetic routes shown in eqn (3) and (4) in [Scheme 1](#sch1){ref-type="fig"}.^[@cit11]^ The carbene **B**, likely highly electrophilic, could then undergo an electrophilic cyclization to yield the final 2-aminopyrroles **3**, thus constituting a gold-catalyzed formal \[3+2\] cycloaddition ([Scheme 1](#sch1){ref-type="fig"}, eqn (2)). Herein, we report the successful implementation of this mechanistic design to a facile and practical synthesis of a wide range of polysubstituted 2-aminopyrroles, which are common structural motifs found in natural products and pharmacologically active molecules ([Fig. 1](#fig1){ref-type="fig"})^[@cit12]^ and are difficult to access *via* traditional methods for pyrrole synthesis.^[@cit13]^ Most importantly, an α-imino gold carbene is most likely generated as the key intermediate on the basis of both mechanistic studies and theoretical calculations, thereby providing a strategically-novel, atom-economic route to the generation of gold carbenes. ![Synthetic design for the atom-economic generation of α-imino gold carbenes: formation of 2-aminopyrroles **3** through gold-catalyzed formal \[3+2\] cycloaddition between ynamides **1** and isoxazoles **2**.](c4sc02596b-s1){#sch1} {#fig1} Results and discussion ====================== At the outset, ynamide **1a** and 3,5-dimethylisoxazole **2a** were used as the reacting species and a series of experiments were performed in order to validate our approach. To our delight, the expected product **3a** was indeed formed in 70% ^1^H NMR yield in the presence of 5 mol% IPrAuNTf~2~ ([Table 1](#tab1){ref-type="table"}, entry 1). Then, various typical gold catalysts with a range of electronic and steric characteristics were screened ([Table 1](#tab1){ref-type="table"}, entries 2--7), and (**Ar**O)~3~PAuNTf~2~ (**Ar** = 2,4-di-*tert*-butylphenyl) gave the best yield of the desired product ([Table 1](#tab1){ref-type="table"}, entry 7). Somewhat surprisingly, AgNTf~2~ could also catalyze this reaction in 50% yield ([Table 1](#tab1){ref-type="table"}, entry 8). Notably, without a metal catalyst, the reaction failed to give even a trace of **3a**, and PtCl~2~ and Zn(OTf)~2~ were not effective in promoting this reaction ([Table 1](#tab1){ref-type="table"}, entries 9--10).^[@cit14]^ The reaction proved to be less efficient when it was performed at a reduced temperature ([Table 1](#tab1){ref-type="table"}, entry 11). In addition, the use of 2 equiv. of **2a** also gave the desired pyrrole **3a** in 90% yield ([Table 1](#tab1){ref-type="table"}, entry 12). ###### Optimization of reaction conditions[^*a*^](#tab1fna){ref-type="fn"} {#ugr2} ------------------------------------------ -------------------------------------------------------------- --------------------- ----- 1 IPrAuNTf~2~ DCE, 80 °C, 3 h 70 2 Ph~3~PAuNTf~2~ DCE, 80 °C, 3 h 69 3 Et~3~PAuNTf~2~ DCE, 80 °C, 3 h 54 4 Cy-JohnPhosAuNTf~2~ DCE, 80 °C, 3 h 71 5 BrettPhosAuNTf~2~ DCE, 80 °C, 12 h 27 6 Au(III)[^*c*^](#tab1fnc){ref-type="table-fn"} DCE, 80 °C, 3 h 34 7 (**Ar**O)~3~PAuNTf~2~ [^*d*^](#tab1fnd){ref-type="table-fn"} DCE, 80 °C, 3 h 95 8 AgNTf~2~ DCE, 80 °C, 3 h 50 9[^*e*^](#tab1fne){ref-type="table-fn"} PtCl~2~ toluene, 80 °C, 3 h \<5 10[^*e*^](#tab1fne){ref-type="table-fn"} Zn(OTf)~2~ (10 mol%) DCE, 80 °C, 3 h \<5 11 (**Ar**O)~3~PAuNTf~2~ [^*d*^](#tab1fnd){ref-type="table-fn"} DCE, 60 °C, 5 h 75 12[^*f*^](#tab1fnf){ref-type="table-fn"} (**Ar**O)~3~PAuNTf~2~ [^*d*^](#tab1fnd){ref-type="table-fn"} DCE, 80 °C, 3 h 90 ^*a*^Reaction conditions: \[**1a**\] = 0.05 M; DCE = 1,2-dichloroethane. ^*b*^Measured by ^1^H NMR using diethyl phthalate as the internal standard. ^*c*^Dichloro(2-picolinato)gold([iii]{.smallcaps}). ^*d*^ **Ar** = 2,4-di-*tert*-butylphenyl. ^*e*^ **1a** was decomposed. ^*f*^2.0 equiv. of **2a** was used. With the optimized reaction conditions in hand, the scope of the transformation was explored. As seen from the results collected in [Table 2](#tab2){ref-type="table"}, the reaction proceeded smoothly with various ynamide substrates **1**, and the yields ranged from 58% to 96%. For example, ynamides with different protecting groups, even the Ns group ([Table 2](#tab2){ref-type="table"}, entries 4--5), readily gave the desired 2-aminopyrroles **3a--f** ([Table 2](#tab2){ref-type="table"}, entries 1--6). Of note, an excellent yield could be achieved in the case of an ynamide with an oxazolidinone moiety and no dimerization reaction was observed ([Table 2](#tab2){ref-type="table"}, entry 6).^[@cit15]^ When R^1^ is an allyl group, the desired **3j** could also be formed in 86% yield, and no cyclopropanation product was formed ([Table 2](#tab2){ref-type="table"}, entry 10).^[@cit16],[@cit5g]^ Other aryl-substituted ynamides were also suitable substrates for this reaction, giving the corresponding functionalized pyrroles **3k--l** in excellent yields ([Table 2](#tab2){ref-type="table"}, entries 11--12). Interestingly, for styryl or cyclopropyl-substituted ynamides, this reaction still led to 75% yield and 58% yield, respectively ([Table 2](#tab2){ref-type="table"}, entries 13--14). The molecular structure of **3a** was further confirmed by X-ray diffraction ([Fig. 2](#fig2){ref-type="fig"}).^[@cit17]^ ###### Reaction scope for different ynamides **1** [^*a*^](#tab2fna){ref-type="fn"} {#ugr3} ------------------------------- {#ugr4} ^*a*^Reactions run in vials; \[**1**\] = 0.05 M; isolated yields are reported. {#fig2} We next extended the reaction to different 3,5-disubstituted isoxazoles **2**. To our delight, the reaction of ynamide **1i** with various isoxazole substrates **2** worked well under the above optimized reaction conditions, giving versatile polysubstituted 2-aminopyrroles **3o--z** in generally good to excellent yields. As summarized in [Table 3](#tab3){ref-type="table"}, a range of aryl-substituted isoxazoles **2c--g** were successful (R^2^ = aryl), delivering the desired **3p--t** in 72--96% yield ([Table 3](#tab3){ref-type="table"}, entries 2--6). In addition, when R^1^ is an aryl group, the reaction also worked well to afford the corresponding pyrroles **3v--w** in excellent yields ([Table 3](#tab3){ref-type="table"}, entries 8--9). Pleasingly, methyl 3-pyrrolecarboxylate **3x** was formed in 90% yield from the corresponding isoxazole ([Table 3](#tab3){ref-type="table"}, entry 10). It should be mentioned that 3-formylpyrroles **3y--z** could also be prepared in serviceable yields ([Table 3](#tab3){ref-type="table"}, entries 11--12). ###### Reaction scope for different isoxazoles **2** [^*a*^](#tab3fna){ref-type="fn"} {#ugr5} ------------------------------- {#ugr6} ^*a*^Reactions run in vials; \[**1i**\] = 0.05 M; isolated yields are reported. Interestingly, when the scope of the method was extended to fully-substituted isoxazoles **4**, the reaction also proceeded well, allowing the convenient synthesis of deacylated polysubstituted 2-aminopyrroles **5**. A series of readily-available substituted ynamides was first examined. The corresponding pyrroles **5a--d** were obtained in 72--85% yield ([Table 4](#tab4){ref-type="table"}, entries 1--4). Then, isoxazoles **4** with substituents at the 4-position were also investigated, giving the products **5e--m** in mostly good to excellent yields ([Table 4](#tab4){ref-type="table"}, entries 5--13). Notably, methyl 3-pyrrolecarboxylate **5n** could also be obtained in 77% yield from the corresponding 4-substituted isoxazole, which is complementary to the above protocol based on the 3,5-disubstituted isoxazoles **2** ([Table 4](#tab4){ref-type="table"}, entry 14 *vs.* [Table 3](#tab3){ref-type="table"}, entry 10). In particular, the 3,4-diphenyl substituted isoxazole also reacted smoothly, delivering the 3,4,5-triphenyl substituted pyrrole **5o** in a respectable 60% yield ([Table 4](#tab4){ref-type="table"}, entry 15). ###### Reaction scope for different ynamides **1** and 4-substituted isoxazoles **4** [^*a*^](#tab4fna){ref-type="fn"} {#ugr7} ------------------------------- {#ugr8} ^*a*^Reactions run in vials; \[**1**\] = 0.05 M; isolated yields are reported. To further test the practicality of the current catalytic system, a gram-scale reaction of 1.36 g of **1a** and 1.07 g of **2a** was carried out with a much lower catalyst loading (1 mol%), and 1.72 g of the desired pyrrole **3a** was formed in 85% yield, highlighting the synthetic utility of this chemistry (eqn (5)). Interestingly, the reaction could also be performed well even in water to afford the desired product **3a** in 80% yield and no hydration of the ynamide was observed (eqn (6)),^[@cit10a]--[@cit10c]^ thus making this protocol more practical and environmentally benign. This chemistry can also be used to construct *N*-heteropyrrolizines, which are present in a variety of bioactive molecules.^[@cit18],[@cit12k]^ For example, treatment of ynamide **1p** with isoxazole **4a** under the optimized reaction conditions gave the pyrrole **5p**, which could be converted into fused 2-aminopyrrole **6** in basic conditions in a one-pot process (63% two-step overall yield, eqn (7)). Compound **6** might serve as a precursor for the synthesis of lipoxygenase inhibitors ([Fig. 1](#fig1){ref-type="fig"}).^[@cit12k]^ The sulfonamide could be readily transformed into a free amine ([Scheme 2](#sch2){ref-type="fig"}). For example, the reaction of ynamide **1d** with isoxazole **2c** under the optimized reaction conditions furnished pyrrole **7** in 81% yield. Nitrogen protection of **7** with a methyl group and subsequent removal of the Ns group using the standard conditions (PhSH, K~2~CO~3~) resulted in the formation of species **7a** (53% two-step overall yield). Subsequent deprotection of the benzyl group in **7a** could be realized by performing MnO~2~-mediated oxidation followed by hydrolysis to afford **7b** in 56% yield.^[@cit13e]^ {#sch2} To probe the mechanism of this reaction, we first synthesized the alkyl-substituted ynamide **1q** as the alkyl-substituted gold carbene is well-known in the gold-catalyzed cycloisomerizations of enynes; \[1,2\] hydride shift followed by elimination of the gold catalyst was involved as the critical deauration step.^[@cit1e],[@cit19]^ Indeed, as depicted in eqn (8), when ynamide **1q** reacted with **2a** under the standard reaction conditions, none of the desired pyrrole was detected and α,β-unsaturated amide **8** was isolated in 25% yield. Amide **8** is supposed to be derived from \[1,2\] hydride shift followed by elimination of the gold catalyst and subsequent hydrolysis. This result indicated that a gold carbene is most likely generated as the key intermediate in this process. On the other hand, the low chemoselectivity in the case of *n*-butyl substituted ynamide shows the importance of aryl substituents on the ynamides to keep a high reactivity for the reactions in [Tables 2--4](#tab2 tab3 tab4){ref-type="table"}.^[@cit20]^ In addition, it was found that a key intermediate 3*H*-pyrrole **5l′** could be detected and isolated in the case of the reaction of ynamide **1i** with fully-substituted isoxazole **4i** ([Table 4](#tab4){ref-type="table"}, entry 12). To further demonstrate this process, we monitored the reaction by ^1^H NMR spectroscopy, as depicted in [Fig. 3](#fig3){ref-type="fig"}. Here, the reaction was performed in the presence of 2 mol% (**Ar**O)~3~PAuNTf~2~ in CDCl~3~ in order to better track the reaction intermediates. At the early stage of the reaction, we could clearly observe the formation of the 3*H*-pyrrole **5l′**, which was gradually transformed into the final 1*H*-pyrrole **5l**. {#fig3} A plausible mechanism to rationalize the formation of pyrrole **3** or **5** is illustrated in [Scheme 3](#sch3){ref-type="fig"}, in light of the above experimental observations and density functional theory (DFT) computations (see ESI[†](#fn1){ref-type="fn"} for details).^[@cit21]^ Initially, nucleophilic attack of isoxazole **2** or **4** to the Au([i]{.smallcaps})-ligated alkyne of ynamide **1** forms vinyl gold intermediate **A** by overcoming a moderate barrier (12.2 kcal mol^--1^). Intermediate **A** isomerizes into the gold carbene intermediate **B** upon breakage of the isoxalic N--O bond,^[@cit22]^ again requiring an activation energy around 12.0 kcal mol^--1^. Subsequent 1,5-cyclization^[@cit23]^ within intermediate **B** readily occurs to afford the Au([i]{.smallcaps})-ligated 3*H*-pyrrole **C**, which upon ligand exchange with another ynamide **1** releases 3*H*-pyrrole **D**. The whole process is highly exothermic with free energy release amounting to 52 kcal mol^--1^. For 3,5-disubstituted isoxazoles **2**, 3*H*-pyrrole **D** readily isomerizes into the final aromatic 1*H*-pyrrole **3** by sigmatropic H-migrations.^[@cit24]^ In the case of fully-substituted isoxazole substrates **4**, **D** is ultimately transformed into the final 1*H*-pyrrole **5**, presumably by a water-assisted deacylative aromatization.^[@cit25]^ {ref-type="table"}, entry 7): relative free energies (Δ*G* ~sol~, in kcal mol^--1^) of key intermediates and transition states were computed at the M06/6-31+G(d)/SDD level in 1,2-dichloroethane at 298 K.](c4sc02596b-s3){#sch3} Conclusions =========== In summary, we have developed a novel gold-catalyzed formal \[3+2\] cycloaddition between ynamides and isoxazoles, leading to the concise and flexible synthesis of polysubstituted 2-aminopyrroles. This methodology makes it possible to introduce four substituents onto a pyrrole ring very freely with high efficiency. Of particular interest, fully substituted isoxazoles also react under deacylation, closing a further gap in the reaction scope. Moreover, an α-imino gold carbene is the most likely intermediate based on both mechanistic studies and theoretical calculations, thus providing a new strategy for the generation of gold carbenes, especially in an atom-economic way. Studies to elucidate the detailed mechanism and further synthetic applications of the current protocol are in progress in our laboratory. We are grateful for the financial support from NSFC (no. 21102119, 21273177 and 21272191), RFDP (20130121110004), NFFTBS (no. J1310024), and the Program for Changjiang Scholars and Innovative Research Team in University. [^1]: †Electronic supplementary information (ESI) available. CCDC [1009902](1009902). For ESI and crystallographic data in CIF or other electronic format see DOI: [10.1039/c4sc02596b](10.1039/c4sc02596b) Click here for additional data file. Click here for additional data file.

Q: Подскажите как выполнить проверку даты В постах проставляется начальная и конечная дата, если текущая дата попадает в этот диапазон - посты выводятся, если не попадает - выводится счетчик. Есть вот такой код: Дату вывода постов я получаю в формате 20171023 в счетчике она должна быть в формате October 27, 2017 00:00:00, получится ли у меня её преобразовать или лучше сразу получать её в таком формате? Как получить ближайшую дату, что бы передать её в счётчик? Нужно проверить даты всех постов и выбрать самую ближайшую. <?php $today = date('Ymd'); $args = array( 'post_type' => 'flash-sale', 'posts_per_page' => -1, 'meta_query' => array( array( 'key' => 'start_sales', 'compare' => '<=', 'value' => $today, ), array( 'key' => 'end_sales', 'compare' => '>=', 'value' => $today, ) ), ); $the_query = new WP_Query( $args ); ?> <?php if ( $the_query->have_posts() ) : ?> <?php while ( $the_query->have_posts() ) : $the_query->the_post(); ?> <?php the_title('<h3>','</h3>'); ?> <?php the_content(); ?> <?php endwhile; ?> <?php wp_reset_postdata(); ?> <?php else : ?> <div class="countdown-sale"> <div class="countdown" data-date="October 27, 2017 00:00:00"></div> <p>Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus rhoncus sapien ut ipsum condimentum, sit amet pellentesque lorem vulputate.</p> <script type="text/javascript"> jQuery(function() { var endDate = "December 1, 2017 00:00:00"; jQuery('.countdown').countdown({ date: endDate }); }); </script> </div> <?php endif; ?> A: Вопрос решен, мой вариант решения ниже: <?php $today = date('Ymd'); $args = array( 'post_type' => 'flash-sale', 'posts_per_page' => -1, 'meta_query' => array( array( 'key' => 'start_sales', 'compare' => '<=', 'value' => $today, ), array( 'key' => 'end_sales', 'compare' => '>=', 'value' => $today, ) ), ); $the_query = new WP_Query( $args ); ?> <?php if ( $the_query->have_posts() ) : ?> <?php while ( $the_query->have_posts() ) : $the_query->the_post(); ?> <?php the_title('<h3>','</h3>'); ?> <?php the_content(); ?> <?php endwhile; ?> <?php wp_reset_postdata(); ?> <?php else : ?> <?php $today = date('Ymd'); $args = array( 'post_type' => 'flash-sale', 'posts_per_page' => 1, 'orderby' => 'meta_value', 'meta_key' => 'start_sales', 'order' => 'ASC', 'meta_query' => array( array( 'key' => 'start_sales', 'compare' => '>=', 'value' => $today, ) ), ); $the_query = new WP_Query( $args ); ?> <?php if ( $the_query->have_posts() ) { while ( $the_query->have_posts() ) { $the_query->the_post(); ?> <?php $feature_sale = get_post_meta( get_the_ID(), 'start_sales' ); $date = strtotime($feature_sale[0]); $new_date = date('F j, Y H:i:s', $date); ?> <div class="countdown-sale"> <div class="countdown" data-date="<?php echo $new_date ?>"></div> <p>Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus rhoncus sapien ut ipsum condimentum, sit amet pellentesque lorem vulputate.</p> <button class="button alt modal-trigger" data-modal="subscribe-modal">Subscribe</button> <div class="modal-overlay" id="subscribe-modal"> <div class="modal-container"> <div class="holder"> <div class="modal"> <div class="close icon-close"></div> <div class="smart-forms"> <div class="form-body"> <?php gravity_form(3, false, false, false, '', true, 12); ?> </div> </div> </div> </div> </div> </div> <script type="text/javascript"> jQuery(function() { var endDate = "<?php echo date('F j, Y') . ' 00:00:00' ?>"; jQuery('.countdown').countdown({ date: endDate, onEnd: function() { jQuery(this.el).addClass('ended'); } }); }); </script> </div> <?php } ?> <?php wp_reset_postdata(); ?> <?php } ?> <?php endif; ?> Здесь идет выборка постов, которые попадают в диапазон дат start_sales и end_sales. Формат даты как я и писал 20171023 <?php $today = date('Ymd'); $args = array( 'post_type' => 'flash-sale', 'posts_per_page' => -1, 'meta_query' => array( array( 'key' => 'start_sales', 'compare' => '<=', 'value' => $today, ), array( 'key' => 'end_sales', 'compare' => '>=', 'value' => $today, ) ), ); $the_query = new WP_Query( $args ); ?> Если посты не попадают в этот диапазон, идет сортировка постов по meta_value. Таким образом я получаю сортировку постов по возрастанию и получаю дату ближайшего поста. Это необходимо для того чтоб передать текущую дату и дату поста в countdown. <?php $today = date('Ymd'); $args = array( 'post_type' => 'flash-sale', 'posts_per_page' => 1, 'orderby' => 'meta_value', 'meta_key' => 'start_sales', 'order' => 'ASC', 'meta_query' => array( array( 'key' => 'start_sales', 'compare' => '>=', 'value' => $today, ) ), ); $the_query = new WP_Query( $args ); ?> Но так как для countdown используется готовая библиотека, нужно преобразовать это в нужный формат даты October 27, 2017 00:00:00. Здесь я получаю нужный формат даты, который передаю в атрибут data-date="" <div class="countdown" data-date="<?php echo $new_date ?>"></div> <?php $feature_sale = get_post_meta( get_the_ID(), 'start_sales' ); $date = strtotime($feature_sale[0]); $new_date = date('F j, Y H:i:s', $date); ?> Ну и наконец я указываю в скрипте текущую дату, также приведенную в нужный формат var endDate = "<?php echo date('F j, Y') . ' 00:00:00' ?>"; P.S.: понимаю, что код можно было сделать красивее, но это мое быстрое решение. Если есть варианты - предлагайте

Scandal-hit Italy take on Spain at Euro 2012 10 June 2012 09:16 Italy are ready to consign their turbulent Euro 2012 preparation to the history books on Sunday when they take on holders Spain in their Group C opener. A match-fixing scandal back home and a 3-0 hammering by Russia in a friendly, their third defeat on the bounce, has seen the Azzurri arrive at these championships in disarray. Even so, Italy midfielder Daniele De Rossi, set to play as a libero against the Spaniards in a revamped formation, believes that although the fans may feel disengaged now, they will soon get behind the team and help them put their troubles to bed. "That could be the case (that the fans feel distant) given what's happened over the last few months but all will be forgotten, not least because those playing (for Italy) have nothing to do with what has happened," he said. "Just one month before we won the World Cup (in 2006) there was a similar atmosphere. "It is a little annoying because now next to the page about the national team you read about a sporting trial and we can't wait for it to be over. "But it's not a case of the whole footballing world being dirty as things seem, the national team is above that." Spain and Barcelona centre-back Gerard Pique believes that will only make the Italians stronger, as it has in the past. Already in 2006 they won the World Cup directly following a match-fixing scandal and they did so too in 1982 just two years after yet another sordid affair. "It seems Italy have been destabilised and have their heads elsewhere. But those are the circumstances in which they are at their strongest," insisted Pique. "I am convinced that Italy will be a very complicated opponent to play against and that they will qualify for the quarters." Spain are not without their own concerns as hot-shot David Villa misses the tournament as he recovers from a broken leg. Villa - Spain's all-time record goalscorer - was Spain's main hero two years ago when they lifted the World Cup, but despite his absence Spain winger Pedro believes the title-holders will not want for goals. "David Villa is a very important player due to the goals he scores, he's a very dangerous forward. "But in the squad there are players just behind who are good finishers like (David) Silva, (Andres) Iniesta, (Juan) Mata or (Jesus) Navas. "We'll have to make up for all the goals David scores amongst the rest of us to go as far as possible in the Euros." Spain are not just missing Villa as centre-back Carles Puyol is also out of the tournament with injury. But Italy midfielder Thiago Motta says those who will come in to replace them are just as good. "They won't be weaker, they're in a very good moment and over the last four years they've won everything," said the Brazlian-born Paris St Germain player. "They're missing players like Villa and Puyol but others can come in and play at the same level. "We need to concentrate on ourselves because they haven't changed much in those four years. "The important thing is that we do ourselves justice on the pitch and play with the desire we've shown from the beginning to win this game." Four years ago Spain were still the nearly men, the great under-achievers who despite generation after generation of great players, never managed to win anything. That all changed when they finally landed the Euro title in Vienna four years ago. Italy centre-back Giorgio Chiellini suggested it was Spain's penalty shoot-out victory over the Azzurri in the quarter-finals that gave them the self-belief to become a great team. Four years on and world champions Spain are undoubtedly the best team on the planet and the clear favourites to prevail again. However, last August Italy actually beat Spain 2-1 in a friendly in Bari, although the world champions largely dominated the match.

Uwe Freiler Uwe Freiler (born 2 May 1966 in Limbach (Saarland)) is a retired German football player. He spent four seasons in the Bundesliga with FC 08 Homburg and SV Waldhof Mannheim. References External links Category:1966 births Category:Living people Category:People from Saarlouis (district) Category:German footballers Category:Association football forwards Category:Bundesliga players Category:2. Bundesliga players Category:FC 08 Homburg players Category:SV Waldhof Mannheim players Category:Footballers from Saarland Category:FK Pirmasens players

# The MB (Meta-Build wrapper) user guide [TOC] ## Introduction `mb` is a simple python wrapper around the GYP and GN meta-build tools to be used as part of the GYP->GN migration. It is intended to be used by bots to make it easier to manage the configuration each bot builds (i.e., the configurations can be changed from chromium commits), and to consolidate the list of all of the various configurations that Chromium is built in. Ideally this tool will no longer be needed after the migration is complete. For more discussion of MB, see also [the design spec](design_spec.md). ## MB subcommands ### `mb analyze` `mb analyze` is responsible for determining what targets are affected by a list of files (e.g., the list of files in a patch on a trybot): ``` mb analyze -c chromium_linux_rel //out/Release input.json output.json ``` Either the `-c/--config` flag or the `-m/--master` and `-b/--builder` flags must be specified so that `mb` can figure out which config to use. The first positional argument must be a GN-style "source-absolute" path to the build directory. The second positional argument is a (normal) path to a JSON file containing a single object with the following fields: * `files`: an array of the modified filenames to check (as paths relative to the checkout root). * `test_targets`: an array of (ninja) build targets that needed to run the tests we wish to run. An empty array will be treated as if there are no tests that will be run. * `additional_compile_targets`: an array of (ninja) build targets that reflect the stuff we might want to build *in addition to* the list passed in `test_targets`. Targets in this list will be treated specially, in the following way: if a given target is a "meta" (GN: group, GYP: none) target like 'blink_tests' or or even the ninja-specific 'all' target, then only the *dependencies* of the target that are affected by the modified files will be rebuilt (not the target itself, which might also cause unaffected dependencies to be rebuilt). An empty list will be treated as if there are no additional targets to build. Empty lists for both `test_targets` and `additional_compile_targets` would cause no work to be done, so will result in an error. * `targets`: a legacy field that resembled a union of `compile_targets` and `test_targets`. Support for this field will be removed once the bots have been updated to use compile_targets and test_targets instead. The third positional argument is a (normal) path to where mb will write the result, also as a JSON object. This object may contain the following fields: * `error`: this should only be present if something failed. * `compile_targets`: the list of ninja targets that should be passed directly to the corresponding ninja / compile.py invocation. This list may contain entries that are *not* listed in the input (see the description of `additional_compile_targets` above and [design_spec.md](the design spec) for how this works). * `invalid_targets`: a list of any targets that were passed in either of the input lists that weren't actually found in the graph. * `test_targets`: the subset of the input `test_targets` that are potentially out of date, indicating that the matching test steps should be re-run. * `targets`: a legacy field that indicates the subset of the input `targets` that depend on the input `files`. * `build_targets`: a legacy field that indicates the minimal subset of targets needed to build all of `targets` that were affected. * `status`: a field containing one of three strings: * `"Found dependency"` (build the `compile_targets`) * `"No dependency"` (i.e., no build needed) * `"Found dependency (all)"` (`test_targets` is returned as-is; `compile_targets` should contain the union of `test_targets` and `additional_compile_targets`. In this case the targets do not need to be pruned). See [design_spec.md](the design spec) for more details and examples; the differences can be subtle. We won't even go into how the `targets` and `build_targets` differ from each other or from `compile_targets` and `test_targets`. The `-b/--builder`, `-c/--config`, `-f/--config-file`, `-m/--master`, `-q/--quiet`, and `-v/--verbose` flags work as documented for `mb gen`. ### `mb audit` `mb audit` is used to track the progress of the GYP->GN migration. You can use it to check a single master, or all the masters we care about. See `mb help audit` for more details (most people are not expected to care about this). ### `mb gen` `mb gen` is responsible for generating the Ninja files by invoking either GYP or GN as appropriate. It takes arguments to specify a build config and a directory, then runs GYP or GN as appropriate: ``` % mb gen -m tryserver.chromium.linux -b linux_rel //out/Release % mb gen -c linux_rel_trybot //out/Release ``` Either the `-c/--config` flag or the `-m/--master` and `-b/--builder` flags must be specified so that `mb` can figure out which config to use. The `--phase` flag must also be used with builders that have multiple build/compile steps (and only with those builders). By default, MB will look for a bot config file under `//ios/build/bots` (see [design_spec.md](the design spec) for details of how the bot config files work). If no matching one is found, will then look in `//tools/mb/mb_config.pyl` to look up the config information, but you can specify a custom config file using the `-f/--config-file` flag. The path must be a GN-style "source-absolute" path (as above). You can pass the `-n/--dryrun` flag to mb gen to see what will happen without actually writing anything. You can pass the `-q/--quiet` flag to get mb to be silent unless there is an error, and pass the `-v/--verbose` flag to get mb to log all of the files that are read and written, and all the commands that are run. If the build config will use the Goma distributed-build system, you can pass the path to your Goma client in the `-g/--goma-dir` flag, and it will be incorporated into the appropriate flags for GYP or GN as needed. If gen ends up using GYP, the path must have a valid GYP configuration as the last component of the path (i.e., specify `//out/Release_x64`, not `//out`). The gyp script defaults to `//build/gyp_chromium`, but can be overridden with the `--gyp-script` flag, e.g. `--gyp-script=gypfiles/gyp_v8`. ### `mb help` Produces help output on the other subcommands ### `mb lookup` Prints what command will be run by `mb gen` (like `mb gen -n` but does not require you to specify a path). The `-b/--builder`, `-c/--config`, `-f/--config-file`, `-m/--master`, `--phase`, `-q/--quiet`, and `-v/--verbose` flags work as documented for `mb gen`. ### `mb validate` Does internal checking to make sure the config file is syntactically valid and that all of the entries are used properly. It does not validate that the flags make sense, or that the builder names are legal or comprehensive, but it does complain about configs and mixins that aren't used. The `-f/--config-file` and `-q/--quiet` flags work as documented for `mb gen`. This is mostly useful as a presubmit check and for verifying changes to the config file. ### `mb gerrit-buildbucket-config` Generates a gerrit buildbucket configuration file and prints it to stdout. This file contains the list of trybots shown in gerrit's UI. The master copy of the buildbucket.config file lives in a separate branch of the chromium repository. Run `mb gerrit-buildbucket-config > buildbucket.config.new && git fetch origin refs/meta/config:refs/remotes/origin/meta/config && git checkout -t -b meta_config origin/meta/config && mv buildbucket.config.new buildbucket.config` to update the file. Note that after committing, `git cl upload` will not work. Instead, use `git push origin HEAD:refs/for/refs/meta/config` to upload the CL for review. ## Isolates and Swarming `mb gen` is also responsible for generating the `.isolate` and `.isolated.gen.json` files needed to run test executables through swarming in a GN build (in a GYP build, this is done as part of the compile step). If you wish to generate the isolate files, pass `mb gen` the `--swarming-targets-file` command line argument; that arg should be a path to a file containing a list of ninja build targets to compute the runtime dependencies for (on Windows, use the ninja target name, not the file, so `base_unittests`, not `base_unittests.exe`). MB will take this file, translate each build target to the matching GN label (e.g., `base_unittests` -> `//base:base_unittests`, write that list to a file called `runtime_deps` in the build directory, and pass that to `gn gen $BUILD ... --runtime-deps-list-file=$BUILD/runtime_deps`. Once GN has computed the lists of runtime dependencies, MB will then look up the command line for each target (currently this is hard-coded in [mb.py](https://code.google.com/p/chromium/codesearch?q=mb.py#chromium/src/tools/mb/mb.py&q=mb.py%20GetIsolateCommand&sq=package:chromium&type=cs)), and write out the matching `.isolate` and `.isolated.gen.json` files. ## The `mb_config.pyl` config file The `mb_config.pyl` config file is intended to enumerate all of the supported build configurations for Chromium. Generally speaking, you should never need to (or want to) build a configuration that isn't listed here, and so by using the configs in this file you can avoid having to juggle long lists of GYP_DEFINES and gn args by hand. `mb_config.pyl` is structured as a file containing a single PYthon Literal expression: a dictionary with three main keys, `masters`, `configs` and `mixins`. The `masters` key contains a nested series of dicts containing mappings of master -> builder -> config . This allows us to isolate the buildbot recipes from the actual details of the configs. The config should either be a single string value representing a key in the `configs` dictionary, or a list of strings, each of which is a key in the `configs` dictionary; the latter case is for builders that do multiple compiles with different arguments in a single build, and must *only* be used for such builders (where a --phase argument must be supplied in each lookup or gen call). The `configs` key points to a dictionary of named build configurations. There should be an key in this dict for every supported configuration of Chromium, meaning every configuration we have a bot for, and every configuration commonly used by developers but that we may not have a bot for. The value of each key is a list of "mixins" that will define what that build_config does. Each item in the list must be an entry in the dictionary value of the `mixins` key. Each mixin value is itself a dictionary that contains one or more of the following keys: * `gyp_crosscompile`: a boolean; if true, GYP_CROSSCOMPILE=1 is set in the environment and passed to GYP. * `gyp_defines`: a string containing a list of GYP_DEFINES. * `gn_args`: a string containing a list of values passed to gn --args. * `mixins`: a list of other mixins that should be included. * `type`: a string with either the value `gyp` or `gn`; setting this indicates which meta-build tool to use. When `mb gen` or `mb analyze` executes, it takes a config name, looks it up in the 'configs' dict, and then does a left-to-right expansion of the mixins; gyp_defines and gn_args values are concatenated, and the type values override each other. For example, if you had: ``` { 'configs`: { 'linux_release_trybot': ['gyp_release', 'trybot'], 'gn_shared_debug': None, } 'mixins': { 'bot': { 'gyp_defines': 'use_goma=1 dcheck_always_on=0', 'gn_args': 'use_goma=true dcheck_always_on=false', }, 'debug': { 'gn_args': 'is_debug=true', }, 'gn': {'type': 'gn'}, 'gyp_release': { 'mixins': ['release'], 'type': 'gyp', }, 'release': { 'gn_args': 'is_debug=false', } 'shared': { 'gn_args': 'is_component_build=true', 'gyp_defines': 'component=shared_library', }, 'trybot': { 'gyp_defines': 'dcheck_always_on=1', 'gn_args': 'dcheck_always_on=true', } } } ``` and you ran `mb gen -c linux_release_trybot //out/Release`, it would translate into a call to `gyp_chromium -G Release` with `GYP_DEFINES` set to `"use_goma=true dcheck_always_on=false dcheck_always_on=true"`. (From that you can see that mb is intentionally dumb and does not attempt to de-dup the flags, it lets gyp do that). ## Debugging MB By design, MB should be simple enough that very little can go wrong. The most obvious issue is that you might see different commands being run than you expect; running `'mb -v'` will print what it's doing and run the commands; `'mb -n'` will print what it will do but *not* run the commands. If you hit weirder things than that, add some print statements to the python script, send a question to gn-dev@chromium.org, or [file a bug](https://crbug.com/new) with the label 'mb' and cc: dpranke@chromium.org.

/* * Copyright 2002-2019 the original author or authors. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * https://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package org.springframework.integration.xml.source; import java.io.File; import java.io.StringReader; import javax.xml.parsers.DocumentBuilder; import javax.xml.parsers.DocumentBuilderFactory; import javax.xml.parsers.ParserConfigurationException; import javax.xml.transform.Source; import javax.xml.transform.dom.DOMSource; import org.w3c.dom.Document; import org.xml.sax.InputSource; import org.springframework.messaging.MessagingException; import org.springframework.util.Assert; import org.springframework.xml.DocumentBuilderFactoryUtils; /** * {@link SourceFactory} implementation which supports creation of a {@link DOMSource} * from a {@link Document}, {@link File} or {@link String} payload. * * @author Jonas Partner * @author Mark Fisher * @author Artem Bilan */ public class DomSourceFactory implements SourceFactory { private final DocumentBuilderFactory documentBuilderFactory; public DomSourceFactory() { this(DocumentBuilderFactoryUtils.newInstance()); this.documentBuilderFactory.setNamespaceAware(true); } public DomSourceFactory(DocumentBuilderFactory documentBuilderFactory) { Assert.notNull(documentBuilderFactory, "'documentBuilderFactory' must not be null."); this.documentBuilderFactory = documentBuilderFactory; } public Source createSource(Object payload) { Source source = null; if (payload instanceof Document) { source = createDomSourceForDocument((Document) payload); } else if (payload instanceof String) { source = createDomSourceForString((String) payload); } else if (payload instanceof File) { source = createDomSourceForFile((File) payload); } if (source == null) { throw new MessagingException("failed to create Source for payload type [" + payload.getClass().getName() + "]"); } return source; } private DOMSource createDomSourceForDocument(Document document) { return new DOMSource(document.getDocumentElement()); } private DOMSource createDomSourceForString(String s) { try { Document document = getNewDocumentBuilder().parse(new InputSource(new StringReader(s))); return new DOMSource(document.getDocumentElement()); } catch (Exception e) { throw new MessagingException("failed to create DOMSource for String payload", e); } } private DOMSource createDomSourceForFile(File file) { try { Document document = getNewDocumentBuilder().parse(file); return new DOMSource(document.getDocumentElement()); } catch (Exception e) { throw new MessagingException("failed to create DOMSource for File payload", e); } } private DocumentBuilder getNewDocumentBuilder() throws ParserConfigurationException { synchronized (this.documentBuilderFactory) { return this.documentBuilderFactory.newDocumentBuilder(); } } }

[Efficacy of walking training in patients with obliterating atherosclerosis of lower limb arteries]. The authors assessed efficacy of treating patients with atherosclerosis obliterans of lower-limb arteries using the method of walking training. Our prospective study included a total of 84 patients suffering from intermittent claudication. They were divided in two groups. The Study Group (Group I) comprised 64 patients undergoing comprehensive treatment including walking training (WT) and the Control Group (Group II) consisted of 20 patients treated without a physical training programme. The patients were followed up monthly during three months, with registration of the pain-free walking distance (PFWD), maximum walking distance (MWD), and the ankle-brachial index (ABI). There were no baseline statistically significant differences between Group I and Group II patients in the parameters examined. The Study Group patients were found to have a considerable increase in the PFWD and MWD values, with the statistical significance of differences from the Control Group patients observed as early as at one month of treatment. The baseline PFWD value in the Study Group amounted to 132.4±14.7 m, being at 3 months 259.7±29.6 m (p<0.05); the MWD value in the Study Group initially equalled 170.2±12.3 m, amounting at 3 months to 364.6±39.2 m (p<0.05). The Study Group patients by the end of follow up were also found to have an increase in the ankle-brachial index by nearly 25% from the baseline level. The carried out treatment resulted in an increase in the number of patients with less severe stages of arterial insufficiency: initially stage IIa was diagnosed in 18 (28.2%) patients and at 3 months in 47 (73.4%) patients, with stage IIb being diagnosed in 46 (71.8 %) and 17 (26.6%) subjects, respectively. A conclusion was drawn that training walking considerably improved clinical manifestations of the disease. Taking into consideration availability of this therapeutic method, it should be recommended for patients suffering from intermittent claudication in long-term courses and in all cases where there are no contraindications thereto.

[Combined ultrasound and roentgen localization in ESWL. Initial clinical experiences]. Interdisciplinary use of extracorporeal shock-wave lithotripsy with sonographic or fluoroscopic guidance has become a routine procedure in clinical practice, for both urolithiasis and cholecystolithiasis. Therefore, newly developed systems with combined locating devices are gaining in importance. A primary sonographically guided lithotripter was extended by a mobile X-ray system. The results were compared with those obtained with a first-generation lithotripter. The results show that sufficient disintegration of stones throughout the urinary tract is possible with both systems. The retreatment rate with both lithotripters was 30% when stone size was comparable. No essential differences in treatment time, shockwave energy and pain were found. The easy localisation of radiolucent stones, convenient positioning of the patient and successful localisation of ureteral stones near to the spine are advantages of the MPL 9000 X. On the other hand, the technically simpler fluoroscopy system and greater ease of electrode changing are advantages of the HM3 Lithotripter. Altogether, differences in the application of the two systems are slight and insignificant except in special cases.

Needed: Tips and tricks for duct testing I am building a new page on our website www.retrotec.com, to help those looking for information on duct testing and adhering to duct testing codes. If anyone out there has tips or tricks that relate to duct testing or duct testing code compliance, please share. Any submission posted on the site will be cited with company name and contact information. Replies to This Discussion SMACNA is typically the universal standard for sheet metal practices of course, but also have the standards for leak testing of any duct system. There are various degrees depending on duct pressure glass, but most residences are not over 2.0" pressure class and most are 1.0" or less. (They would tend to operate at no more than 0.50" WG in any case.) One of my tricks is a manifold for the tester. The plastic interface for the return is too small. The manifold allows me to use less tape for sealing and speeds up the process.I use mdf 1/4 inch. It broke recently I am contemplating what product to use next. Ceiling returns have caused more time and problems than any other facet of duct testing for me. I am still in search of the perfect solution. On ceiling returns i wedge a a paint pole pole to wedge the manifold up against the return to hold it in place. the weight of the hose tends to make this connection difficult. It reduces the loss of bond from tape when powering up. I will use duct tape at times to improve the bond. If it really becomes problematic I run a screw into the ceiling and patch it with caulk before I leave. I carry red rosin paper for large gaps. I bring a rag and cleaner for areas tape does not stick. In problem areas where tape does not want to stick I will lay on strip on the wall first and then tape to that strip I recently purchased a product for vents. Here in California we can seal against the drywall. Title 24 requires the return register box be sealed to the wall. The product called a vent cap. The cost was too high but the complaints I have received over time from my green customers made me bite the bullet. Having used it only a handful of times I am impressed. it improves speed and works well and no more tape. I always power up the fan speed slow and check all my connections before running to a full 25 pascal looking for loose tape and missed registers. There are several methods for duct testing that are allowed here in California. For existing duct residential testing, California allows a duct leakage to outside test. The maximum allowed flow for this test is 10% of the nominal blower capacity. The advantage of this test is that no supply or return registers need to be sealed or taped. You pressurize the duct system at the furnace/ air handler relative to the outside to 25 Pa. This is easy if the furnace/ air handler is in the garage, for example. Then, using a blower door, you pressurize the house to 25 Pa. Since there is no pressure difference across the air supply/ return registers located in the house, none of these need be sealed. The total leakage you read at the duct blaster is duct to outside leakage only. If you don't have access to a blower door, then a total duct leakage test may be done. In this case all supply and return registers need to be sealed. There is a product called duc-bloc that I have used that can speed up the time required to seal the registers. The product creates a seal using a rubber gasket on the sheet rock wall or ceiling around the register. This product is great for hard to reach ceiling or wall registers. I still use painter's tape for floor registers. There are some rules of thumb for the size of duct leaks. Sealing the boots on all registers can reduce duct leakage by as much as 5-10 cfm per boot. Small 2-3 inch x 1/2 inch leaks in the return plenum can leak 75-100 cfm. In the supply plenum near the blower, very small holes or gaps can cause major leaks (> 100 cfm). If you have excess leakage and need to find out where the leaks are then one method is to cut a small hole in the tape of each sealed supply/return register and verify that that duct is pressurized to 25 Pa using another manometer. If it is significantly less, then you have a problem with the duct that feeds that register. I have found completely disconnected ducts using this method. For duct registers in commercial buildings, I plan on making a 2 ft x 2 ft duct "cover" made from the plastic carpet protector product one can find at Lowe's. I'd like to find out what folks thing about using theatrical smoke during duct testing. We've done it once when we were trying to get leakage below 6%. It turned out that the biggest culprit was the new 95% effic. FAU! The smoke made it very easy to find the leakage locations and we'd do it more but I'm concerned that the smoke will damage our duct blaster. We have a Retrotec Q32 - Have I already voided the warrenty? Theatrical smoke is typically made up of diethylene glycol, dipropylene glycol, propylene glycol, or triethylene glycol and glycerin with distilled water. After sustained use you will see a residue build up that can be wiped up. It's up to the manufacturer whether to void a warranty. In regards to theatrical smoke with the Q32, I caution you to make sure not to inject the smoke directly into the fan motor, but instead into the edge of the fan housing. Also, make sure to clean up any residue from the flow sensors, motor and fan housing when the show is over. What I've done for ceiling returns is had my sheet metal shop make a 16"x25" hemed cover(the most common return size here) with a 10.25" hole in the center for the duct blaster coller. I have two 2" pieces of thin all-thread through two holes at each end of the cover& two wingnuts on each. I insert the upper wingnuts through the register & turn them ninety degrees so they catch on the register vanes & then tighten the lower wingnuts. This holds the cover to the register & supports all the weight. When you set your BD up; you run a routine to set the baseline pressure adjustment. Then before removing the rings, turn on the AH and watch the change in the pressure differential WRT outside. A negative change of any size indicates a depressurization, thus the supply side is more leaky then the return side. A positive change of any size indicates a pressurization, thus the return side is more leaky. It doesn't indicate anything about the difference other than it exists. No change is inconclusive. It could be both sides are equally leaky, or almost anything else. Right now, I'm using the Zip Wall poles and custom made pans to seal the registers. Very quiick to seal registers and equipment tear down afterwards. If you plan on doing a lot of duct testing, it makes sense to invest in the set up. I carry the poles in a ski bag which is much cheaper (on Ebay) than buying the Zip Wall bag. Be sure to have at least two 20" poles for those house with registers in the vaulted ceilings. For houses with floor registers, I still use the poles and pans, I just put a new shoe cover over the end to keep from marking up the ceiling. If you're still using blue painter's tape to seal the registers, here's a simple trick. Get a couple of cheap plastic tape dispensers that come with the clear packaging tape. The advantage is that the dispenser has a cutting edge for the tape. It makes taping registers very quick since you use one motion to pull the tape off of the roll and cut it. If the registers are on the ceiling or upper wall, I leave a "tail" on the tape so I can pull off the tape in one piece with a reach extension. It saves having to climb up and down a ladder or step stool. For return registers, I use painter's 12" wide paper to cover the area not covered by the DB hose adapter. It's cheap and easy to cut with a utility knife. For the DB register probe, I use a Dwyer A-303 static pressure probe that has a magnet. I cut the tip off of the probe and bend it at about a 45 degrees to that it slips in between the register blades. No more problems with the probe slipping out. Lastly, if you haven't swapped out your DB duffel bag for one with wheels, you should.

--- abstract: | The Einstein-Podolsky-Rosen argument on quantum mechanics incompleteness is formulated in terms of elements of reality inferred from [*joint*]{} (as opposed to [*alternative*]{}) measurements, in two examples involving entangled states of three spin-$\frac{1}{2}$ particles. The same states allow us to obtain proofs of the incompatibility between quantum mechanics and elements of reality.\ \ PACS numbers: 03.65.Bz author: - 'Adán Cabello[^1]' - | Guillermo García-Alcaine[^2]\ [*Departamento de Física Teórica,*]{}\ [*Universidad Complutense, 28040 Madrid, Spain.*]{} title: 'Quantum mechanics and elements of reality inferred from joint measurements[^3]' --- Introduction ============ The layout of the paper is as follows. In section 2 we briefly review some opinions on a controversial point in Einstein-Podolsky-Rosen’s (EPR) [@EPR35] argument. In section 3 we introduce some notations and a more detailed form of the sufficient condition for joint inference of several elements of reality (ERs) in the same individual system. In section 4 we formulate EPR’s incompleteness argument in terms of ERs inferred from joint measurements (instead of alternative incompatible measurements as in the original paper and all its sequels), in two entangled states of three spin-$\frac{1}{2}$ particles: an extension to three particles of Hardy’s two-particle state [@Hardy93], and GHZ-Mermin state [@GHZ89; @Mermin90a]. In section 5 we show the incompatibility between quantum mechanics (QM) and ERs (Bell’s theorem), in the same entangled states used in section 4. Finally we summarize our conclusions in section 6. Elements of reality and EPR’s incompleteness argument ===================================================== In 1935, EPR presented an argument to prove that QM only provides an “incomplete” description of physical reality [@EPR35]. Their conclusions were striking because of “the very mild character of the sufficient condition for the reality of a physical quantity on which their argument hinged” [@Mermin90b]: “If, without in any way disturbing a system, we can predict with certainty (i.e., with probability equal to unity) the value of a physical quantity, then there exists an element of physical reality corresponding to this physical quantity.” There is a problem, however, in EPR’s argument: “... one cannot make both measurements \[position or momentum of one of the particles in the original EPR example\], and hence both predictions \[position or momentum of the other particle\], simultaneously” [@HB92]. The joint (“simultaneous”) existence in the same individual system of ERs corresponding to two incompatible observables is inferred from the [*possibility*]{} of measuring either of two also mutually incompatible observables[^4]; this fact has been pointed out as “the most significant lacuna” in EPR’s reasoning [@Howard85]. For EPR’s incompleteness argument to be valid [@HB92; @Howard85], [*counterfactual definiteness*]{} (CFD) [@HB92; @Ballentine90] is required, either as an additional assumption [@Howard85], or deduced from locality [@Ballentine90]. Independently of how natural it may appear, in our opinion this spoils EPR’s own dictum: “The elements of the physical reality cannot be determined by [*a priori*]{} philosophical considerations, but must be found by an appeal to results of experiments and measurements” [@EPR35]. In their argument, EPR use the ambiguous expression “can predict” in the condition for existence of ERs in a broad or “weak” sense [@HB92; @CS78; @GHSZ90; @Stapp91; @d'Espagnat93; @Shimony93], meaning that it is [*possible*]{} to make any of the [*alternative*]{} measurements that would provide the data for either inference. But, as it has been pointed out [@HB92; @CS78; @GHSZ90; @Stapp91; @d'Espagnat93; @Shimony93], “can predict” can also be interpreted in a narrow or “strong” sense, meaning that we actually [*do have*]{} sufficient data to predict with certainty the [*concrete*]{} result to which the ER is ascribed. It is a common belief that EPR’s argument does not work if this strong sense is required: “The EPR argument goes through only if “can predict” is understood in the weak sense” [@Shimony93] (see also p 1885 of [@CS78], or p 142 of [@d'Espagnat93]); EPR themselves said: “Indeed, one would not arrive at our conclusion if one insisted that two or more physical quantities can be regarded as simultaneous elements of reality [*only when they can be simultaneously measured or predicted*]{}. On this point of view, since either one or the other, but not both simultaneously, of the quantities [*P*]{} and [*Q*]{} can be predicted, they are not simultaneously real. This makes the reality of [*P*]{} and [*Q*]{} depend upon the process of measurement carried out on the first system, which does not disturb the second system in any way. No reasonable definition of reality could be expected to permit this” [@EPR35]. The main aim of this paper is to show how we can formulate an EPR argument using ERs predicted in the strong sense[^5]. Elements of reality inferred from joint measurements ==================================================== We will call [*strong elements of reality*]{} (SERs) those obtained from EPR’s criterion, taking “can predict” in the strong sense. The name also reflects their [*actuality*]{}, in contradistinction with the ones used in EPR’s paper and in most of the literature: SERs have definite predicted values, instead of an abstract existence without concrete values as the ERs inferred using the weak sense of “can predict”. Note, however, that EPR’s criterion is a [*sufficient*]{} condition, and that the set of assumptions used to infer SERs is in fact [*weaker*]{}: the CFD implicitly used in the original EPR argument is not required[^6]. We will guarantee the clause “without in any way disturbing a system” in EPR’s sufficient condition by requiring that the system to which the ER is assigned be outside the future light cones of the points at which we perform any observation needed to make the prediction. This is how the ERs were inferred in EPR’s original paper: “since at the time of measurement the two systems no longer interact, no real change can take place in the second system in consequence of anything that may be done to the first system” [@EPR35]. More explicitly, we formulate the sufficient condition for existence of a SER in the following way. Let us consider a physical system with two parts $S_1$, $S_2$, and two spacelike separated regions $R_1$, $R_2$ of the respective world tubes. If we can predict with certainty the concrete value of a physical quantity in $R_2$ from the result of a measurement performed in $R_1$, then there is a [*strong element of reality*]{} corresponding to that physical quantity, [*at least*]{} in the part of the world tube of $S_2$ outside the future light cones with vertices in $R_1$ and prior to any external perturbation of $S_2$. The SER need not be the value of the physical quantity itself, as long as it determines univocally such value. For instance, in Bohm’s theory the position of a particle in a Stern-Gerlach apparatus determines the value of the corresponding spin component [@BH93]. Nevertheless, to simplify the discussion we will identify the SER with the value of the physical quantity that it predicts. Our next step is to illustrate when several SERs can be jointly assigned to the same individual system. Suppose that we make a measurement in a region $R_1$ of the world tube of $S_1$, and that the result obtained allows us to infer a SER $r^{(2)}$ for $S_2$ following our previous criterion. On the same individual system, suppose that a second observer makes a measurement in a region $R_2$ of the world tube of $S_2$, spacelike separated from $R_1$, and that his result enables him to infer a SER $r^{(1)}$ for $S_1$ in its originally prepared state (prior to any disturbance). The persistence of the SER $r^{(2)}$ after the measurement at $R_2$ is not guaranteed (in general the measurement disturbs the subsystem $S_2$; it could even destroy it!); the same can be said about the SER $r^{(1)}$ after the measurement at $R_1$. According to the sufficient condition for existence of SERs, we can assign to that individual system two [*joint*]{} SERs, $r^{(1)}$ and $r^{(2)}$, at least in the parts of the world tubes outside the future light cones with vertices in $R_1$ and $R_2$ (loosely speaking, in the parts of the world tubes “previous” to the measurements). For any pair of (point-like) events, one in each region, there is always an inertial reference frame in which both events are simultaneous. Therefore, it is tempting to adopt the usual terminology and denote the joint SERs as “simultaneous” SERs, but we have avoided this in order to emphasize that the combined existence of $r^{(1)}$ and $r^{(2)}$ in the same individual physical system does not depend on any external choice of a reference frame. We need not worry about instantaneity of measurements or simultaneity of three-dimensional sections of the four-dimensional world tubes of $S_1$ and $S_2$, and the generalization to systems with three or more extended spacelike separated parts is immediate: the only requisite on the compound system is that we can make predictions with certainty about properties (possibly non-local) of the individual system in some spacetime regions (prior to any external disturbance), based on measurements performed in other spacelike separated regions. EPR’s argument in terms of joint SER ==================================== We are going to follow EPR’s previously quoted dictum on the necessity of finding ERs “by an appeal to results of experiments”, and reformulate their argument using only ERs jointly inferred from actual measurements (our previously defined joint SERs), without [*a priori*]{} use of alternative inferences. Joint SERs will be obtained for observables that have [*no*]{} common eigenstate; this is a stronger condition than the simple incompatibility; even if they do not have a basis of common eigenstates, two incompatible observables can share some eigenstates; the existence of joint SERs in one of these states will not prove QM incompleteness (of course, $\hat P$, $\hat Q$ have no common eigenstate, but other incompatible observables without this property have been used in the literature). Consider a system of three spin-$\frac{1}{2}$ particles in spacelike separated regions, prepared in the entangled spin state (in the basis of eigenstates of the operators $\hat \sigma _z^{(j)}$, $j=1, 2, 3$), $$\left| \psi \right\rangle = a \left( {\left| {+++} \right\rangle - \left| {+-+} \right\rangle -\left| {-++} \right\rangle } \right) + b \left| {---} \right\rangle\,, \label{uno}$$ with $3\left| a \right|^2 +\left| b \right|^2 = 1$, and $ab\ne 0$. In this state, $$P_\psi \left( {\sigma _x^{(2)}=-1 \left|\,{\sigma _z^{(1)}=+1} \right.} \right) = 1\,, \label{dos}$$ $$P_\psi \left( {\sigma _x^{(1)}=-1 \left|\,{\sigma _z^{(2)}=+1} \right.} \right) = 1\,, \label{tres}$$ where $P_\psi \left( {\sigma _x^{(2)}=-1 \left|\,{\sigma _z^{(1)}=+1} \right.} \right)$ denotes the conditional probability of obtaining the value $\sigma _x^{(2)}=-1$ if $\sigma _z^{(1)}=+1$. If we measure $\hat \sigma _z^{(1)}$ on the first particle and obtain the result $+1$, property (\[dos\]) allows us to assign a SER $\sigma _x^{(2)}=-1$ to the second particle, at least as long as it is not disturbed. Analogously, if a second observer measures $\hat \sigma _z^{(2)}$ on the second particle and obtains the result $+1$, property (\[tres\]) allows him to assign a SER $\sigma _x^{(1)}=-1$ to the first unperturbed particle[^7]. Consider now the non-local (and non-factorizable) observable defined by the projector $$\hat \pi ^{(1+2)} = 1 -\left| {--} \right\rangle \left\langle {--} \right|\,, \label{cuatro}$$ in the system formed by the first and second particles. This observable $\hat \pi ^{(1+2)}$ is compatible with $\hat \sigma _z^{(1)}$ and $\hat \sigma _z^{(2)}$ but not with $\hat \sigma _x^{(1)}$ and $\hat \sigma _x^{(2)}$; in fact, it is easy to check that there is no common eigenstate to the three operators $\hat \sigma _x^{(1)}\otimes 1^{(2)}$, $1^{(1)}\otimes \hat \sigma _x^{(2)}$ and $\hat \pi ^{(1+2)}$. In state (\[uno\]), $$P_\psi \left( {\pi ^{(1+2)}=1 \left|\,{\sigma _z^{(3)}=+1} \right.} \right) = 1\,. \label{cinco}$$ Consequently, if a third observer measures $\hat \sigma _z^{(3)}$ and obtains the result $+1$, he can assign a SER $\pi ^{(1+2)}=1$ to the system formed by the first and second particles. The probability of obtaining the results $\sigma _z^{(1)}=+1$, $\sigma _z^{(2)}=+1$, $\sigma _z^{(3)}=+1$ in a joint measurement of the three observables $\hat \sigma _z^{(j)}$ in spacelike separated regions of the same individual physical system is not zero: $$P_\psi \left( {\sigma _z^{(1)}=+1,\sigma _z^{(2)}=+1,\sigma _z^{(3)}=+1} \right)=\left| a \right|^2\,. \label{seis}$$ In an individual physical system in which these three results are obtained, following (\[dos\]), (\[tres\]) and (\[cinco\]) we can infer three joint SERs: $\sigma _x^{(1)}=-1$, $\sigma _x^{(2)}=-1$ and $\pi ^{(1+2)}=1$. Because there is [*no*]{} common eigenstate of the corresponding observables, according to EPR we would conclude that the quantum state “does not provide a complete description of the physical reality” [@EPR35] of this individual system. We could attempt a similar incompleteness argument in the state of the first two particles given by $$\left| \eta \right\rangle ={1 \over {\sqrt 3}}\left( {\left| {++} \right\rangle -\left| {+-} \right\rangle -\left| {-+} \right\rangle } \right)\,, \label{siete}$$ which is an example of a Hardy state [@Hardy93] that verifies properties (\[dos\]) and (\[tres\]) and is an eigenstate of $\hat \pi ^{(1+2)}$ with eigenvalue 1. But then, the values $\sigma _x^{(1)}=-1$, $\sigma _x^{(2)}=-1$ and $\pi ^{(1+2)}=1$ would not satisfy our condition to be considered joint SERs, because the preparation of the state (\[siete\]), in which the value $\pi ^{(1+2)}=1$ rest, is not spacelike separated from the measurements used to infer the two other values; the measurements of $\hat \sigma _z^{(1)}$, $\hat \sigma _z^{(2)}$ are in the future of the preparation of the state (\[siete\]), and therefore their precise results in an individual system could be influenced by the preparation. As we said in footnote 2, three or more spacelike separated parts are needed in order to meet the condition for existence of joint SERs for incompatible observables (of course, ERs might exist without this [*sufficient*]{} condition being fulfilled, but following EPR, we should not concern ourselves with more precise definitions of the ERs). The third particle in our example (\[uno\]) is a device to allow us the use of the sufficient condition for existence of joint SERs, although the SERs inferred involve only the first two particles; in the next example the three particles play a more symmetrical role. A stronger incompleteness proof can be worked out in the GHZ-Mermin [@GHZ89; @Mermin90a] state of three spin-$\frac{1}{2}$ particles: $$\left| \mu \right\rangle ={1 \over {\sqrt 2}}\left( {\left| {+++} \right\rangle -\left| {---} \right\rangle } \right)\,. \label{ocho}$$ Let us denote $\hat A_1=\hat \sigma _x^{(2)}\otimes \hat \sigma _y^{(3)}$, $\hat A_2=\hat \sigma _y^{(1)}\otimes \hat \sigma _x^{(3)}$, $\hat A_3=\hat \sigma _x^{(1)}\otimes \hat \sigma _y^{(2)}$. In this state we have, $$P_\mu \left( {A_j=\varepsilon _j\left|\,{\sigma _y^{(j)}=\varepsilon _j} \right.} \right)=1\,,\;\;\;\;\varepsilon _j=\pm 1\,,\;\;j=1,2,3\,. \label{nueve}$$ There is no common eigenstate to any two of the observables $\hat A_j$, and nevertheless we can infer joint SER for the three of them, by measuring $\hat \sigma _y^{(j)}$ in spacelike separated regions, whatever the results obtained in these measurements (in our previous example we have to restrict ourselves to those individual physical systems in which $\sigma _y^{(j)}=+1$, $j=1,2,3$). The price to pay for this extension of the argument to all individual systems in state (\[ocho\]) is that the three incompatible observables to which the joint SER are assigned are now non-local (although factorizable), instead of one non-local and two local observables in the previous “probabilistic” example. In both examples (\[uno\]), (\[ocho\]) the incompleteness argument has been formulated in terms of joint SERs, inferred from joint (“simultaneous”, in the usual terminology) local measurements in causally non-connected regions of the same individual system, which was considered impossible until now [@CS78; @d'Espagnat93; @Shimony93]. Bell-EPR theorem in terms of joint SER ====================================== In recent years there have been many proofs of Bell-EPR theorem [@Bell64] “without inequalities” [@Hardy93; @GHZ89; @Mermin90a; @GHSZ90]. Arguments similar to those in the last section, in the same entangled states, lead us to proofs of these kind that are simple variants, using joint SERs, of Hardy’s [@Hardy93] and Mermin’s [@Mermin90a] results. State (\[uno\]) has the following additional properties $$P_\psi \left( {\sigma _z^{(2)}=-1 \left|\,{\sigma _x^{(1)}=+1} \right.} \right) = 1\,, \label{diez}$$ $$P_\psi \left( {\sigma _z^{(1)}=-1 \left|\,{\sigma _x^{(2)}=+1} \right.} \right) = 1\,, \label{once}$$ $$P_\psi \left( {\sigma _x^{(1)}=+1,\sigma _x^{(2)}=+1,\sigma _z^{(3)}=+1} \right)=\frac{1}{4} \left| a \right|^2\,. \label{doce}$$ If we measure the observables $\hat \sigma _x^{(1)}$, $\hat \sigma _x^{(2)}$ and $\hat \sigma _z^{(3)}$, and obtain $+1$ in the three cases (the probability for this is not zero, according to (\[doce\])), properties (\[diez\]) and (\[once\]), together with property (\[cinco\]), allow us to infer respectively three joint SERs associated with the [*compatible*]{} observables $\hat \sigma _z^{(2)}$, $\hat \sigma _z^{(1)}$ and $\hat \pi^ {(1+2)}$, with predicted values $-1$, $-1$ and 1, respectively. But according to QM, in a joint measurement (feasible in principle) of these observables in any state, if the first two results are $\sigma _z^{(1)} = \sigma _z^{(2)} = -1$, the third one will necessarily be $\pi^ {(1+2)} = 0$: QM is not compatible with EPR’s elements of reality, even in their less controversial form (joint SERs). State (\[uno\]) is the tensor product of the Hardy state (\[siete\]) by state $\left| + \right\rangle $ of the third particle, entangled with the product of state $\left| {--} \right\rangle $ of the first two particles by state $\left| - \right\rangle $ of the third one. The presence of the third particle allows us to prove the incompatibility with QM (Bell’s theorem) using three joint SERs that can never be found as results of the corresponding measurements, not only on state (\[uno\]), but on [*any*]{} quantum state; by contrast, Hardy’s two ERs, $\sigma _z^{(1)} = -1$, $\sigma _z^{(2)} = -1$, can never be obtained in a joint measurement in state (\[siete\]), but states in which these results can be obtained do trivially exist. The contradiction between ERs for compatible observables and QM found by Mermin [@Mermin90a] in the GHZ-Mermin state (\[ocho\]) can also be formulated in terms of joint SERs. Let us denote $\hat B_1=\hat \sigma _y^{(2)}\otimes \hat \sigma _y^{(3)}$, $\hat B_2=\hat \sigma _y^{(1)}\otimes \hat \sigma _y^{(3)}$, $\hat B_3=\hat \sigma _y^{(1)}\otimes \hat \sigma _y^{(2)}$; then we have, $$P_\mu \left( {B_j=\varepsilon _j\left|\,{\sigma _x^{(j)}=\varepsilon _j} \right.} \right)=1\,,\;\;\;\;\varepsilon _j=\pm 1\,,\;\;j=1,2,3\,, \label{trece}$$ $$P_\mu \left( {\sigma _x^{(1)}\sigma _x^{(2)}\sigma _x^{(3)}=-1} \right)=1\,. \label{catorce}$$ If the three $\hat \sigma _x^{(j)}$ are measured in spacelike separated regions of the same individual system in state (\[ocho\]), equation (\[trece\]) allows us to infer three joint SERs, $B_j=\varepsilon _j$, that must verify the relation $\varepsilon _1\varepsilon _2\varepsilon _3=-1$ following equation (\[catorce\]). On the other hand, the product of the three [*compatible*]{} observables $\hat B _{j}$ is the unit operator, and therefore there is [*no*]{} quantum state in which the results of their measurement in the same individual system satisfy that relation (each result will be $\pm 1$, but the product of the three is always $+1$). This proves again the incompatibility between QM and joint SERs. Note that all observables used to infer ERs in sections 4 and 5 are (local) spin components, and therefore there would be no difficulty for their joint measurement and the corresponding inference of joint SERs. Non-localities appear only in the observables to which the ERs are assigned; in this sense the experimental check of some of the predictions in sections 4 and 5 could be problematic, but our (gedanken) examples show the differences between QM [*theory*]{} and any theory that includes EPR’s elements of reality: independently of any experimental confirmation of either theory, we have shown that ERs cannot be used to “complete” QM. Summary ======= Working only with what we have called “joint SERs”, inferred from joint measurements in the same individual physical system (a more [*actual*]{} kind than the usual ERs inferred from [*alternative*]{} incompatible measurements), we have reached the following conclusions. \(i) If we accept a very mild sufficient condition for the existence of elements of reality, joint SERs for observables without [*any*]{} common eigenstate will exist. On these premises, QM would be incomplete. This formulation of the EPR argument in terms of joint SERs was considered impossible. \(ii) There are sets of joint SERs for compatible observables that, according to QM, can never be obtained as results of the corresponding joint measurements in [*any*]{} individual system. However plausible, elements of reality are incompatible with QM (Bell’s theorem). Both theorems have been proved using quite similar arguments in the same [*entangled*]{} quantum states (\[uno\]), (\[ocho\]). Acknowledgements ================ We would like to thank Gabriel Álvarez and Emilio Santos for their many helpful comments and the anonymous referees for pointing out to us some obscure points and for their constructive comments. [99]{} Einstein A, Podolsky B and Rosen N 1935 [*Phys. Rev.*]{} [**47**]{} 777 Hardy L 1993 [*Phys. Rev. Lett.*]{} [**71**]{} 1665 Greenberger D M, Horne M A and Zeilinger A 1989 [*Bell’s Theorem, Quantum Theory, and Conceptions of the Universe*]{} ed M Kafatos (Dordrecht: Kluwer) p 69 Mermin N D 1990 [*Phys. Rev. Lett.*]{} [**65**]{} 3373 Mermin N D 1990 [*Boojums all the Way Through*]{} (Cambridge: Cambridge University Press) p 177 Hájek A and Bub J 1992 [*Found. Phys.*]{} [**22**]{} 313 Howard D 1985 [*Stud. Hist. Phil. Sci.*]{} [**16**]{} 171 Redhead M L G 1987 [*Incompleteness, Nonlocality, and Realism*]{} (Oxford: Clarendon) Ballentine L E 1990 [*Quantum Mechanics*]{} (Englewood Cliffs, NJ: Prentice-Hall) p 453 Clauser J F and Shimony A 1978 [*Rep. Prog. Phys.*]{} [**41**]{} 1881 Greenberger D M, Horne M A, Shimony A and Zeilinger A 1990 [*Am. J. Phys.*]{} [**5**]{} 1131 Stapp H P 1991 [*Found. Phys.*]{} [**21**]{} 1 d’Espagnat B 1993 [*Bell’s Theorem and the Foundations of Modern Physics*]{} ed A van der Merwe and F Selleri (Singapore: World Scientific) p 139 Shimony A 1993 [*Search for a Naturalistic World View*]{} vol II (New York: Cambridge University Press) p 187 Pitowsky I 1991 [*Phys. Lett. A*]{} [**156**]{} 137 Bohm D and Hiley B J 1993 [*The Undivided Universe*]{} (London: Routledge) Bell J S 1964 [*Physics*]{} [**1**]{} 195 [^1]: Present address: [*Departamento de Física Aplicada, Universidad de Sevilla, 41012 Sevilla, Spain.*]{} Electronic address: fite1z1@sis.ucm.es [^2]: Electronic address: fite114@sis.ucm.es [^3]: J. Phys. A: Math. Gen. [**30**]{}, 725-732 (1997). [^4]: In [@Redhead87], Redhead discusses how, if the [*minimal instrumentalistic interpretation*]{} of the QM formalism is complemented by any of several possible [*views*]{} (p 45), then “the argument for incompleteness of the QM goes through without any consideration of the alternative possibilities of measuring \[several incompatible observables\]” (p 78). The locality principle and the inference of only one element of reality (in the singlet state, in Redhead’s example) will be enough to prove QM incompleteness (p 77). In this paper we will prove how EPR’s argument on QM incompleteness can be reformulated restricting ourselves to a minimal set of assumptions, both on QM interpretation ([*minimal instrumentalistic interpretation*]{}, in Redhead’s terminology), and on the condition for existence of ERs. [^5]: Such form of the argument will not work for systems with only two spacelike separated parts: in these systems, any couple of joint ERs predicted in the strong sense will correspond to one observable on each of the two spacelike separated parts, that are always compatible; in order to have ERs in the strong sense for incompatible observables we need three or more spacelike separated parts. [^6]: The word [*counterfactual*]{} is used in several contexts: quoting Ballentine [@Ballentine90], “[*Counterfactual definiteness*]{} (abbreviated CFD) ...occurs in the EPR argument when they assert that if we had measured the position of particle 1 we could have learned the position $x_2$ of particle 2, and if we had measured the momentum of particle 1 we could have learned the momentum $p_2$ of particle 2. Although only one of these measurements can actually be carried out in a single case, the conclusion that both values $x_2$ and $p_2$ are well defined in nature is an instance of CFD.” [*Counterfactual*]{} is used also to qualify those predictions that are not actually tested (or even that [*cannot*]{} be tested); in this sense, we subscribe Pitowsky’s opinion [@Pitowsky91]: “This \[the fact that the experiment to test a prediction cannot be conducted\] does not render the prediction invalid. It simply makes it untestable.” We think that counterfactuality in this second sense is less worrying that the one involved in the first, where the [*joint*]{} inference of several ERs is based on [*alternative*]{} incompatible measurements. [^7]: Of course, after measuring $\sigma _z^{(1)}$, $\sigma _z^{(2)}$, the values predicted for $\sigma _x^{(1)}$, $\sigma _x^{(2)}$ cannot be verified, and in this sense these joint SERs are also counterfactual, but at least the measurements needed to infer both values can be made in the same individual system, and joint counterfactual (alternative) inferences are not involved; see footnote 3.

Archive For The “UK” Category Shortly before I left for Australia, I headed to the New Forest for a night at Hotel TerraVina with one of my good friends, Laura. The New Forest is one of my favourite areas in the UK. I love the wild horses, the beautiful scenery and the quaint old pubs, so I was excited to… I visited Northern Ireland in January for the first time, spending a weekend in Belfast as a lovely Christmas present. I had a great time and decided I’d like to see more of the country. So last month returned to Northern Ireland to take a road trip and explore the other cities and towns,… I’m a massive fan of dim sum, so when I got the opportunity to visit Royal China and try out their new dim sum menu, I leapt at the chance. I often used to walk past the Baker Street branch, jealously checking out the people inside but I’ve never actually been in myself. So…

Online Casino Prime News & Updates As the premiere source for gambling news and updates, Online Casino Prime has everything about online casinos and Internet gambling. We offer a complete, detailed and in-depth information with regards to online casino industry. European Operators to Benefit from Legalized Online Casinos in New Jersey Online casino prime providers in the United Kingdom are rejoicing about the recent approval of online gambling in the state of New Jersey. These gambling operators are now keeping their eye on the market with excitement and a positive outlook. Some of them already have a deal in place with the brick and mortar casino owners in the Atlantic City where interactive gaming will be offered from. This includes the Bwin.Party Entertainment that has an existing deal with the MGM company. Bwin.Party’s share prices relatively increased following the announcement of Governor Christie’s approval of casino online in the state that includes table and card games, poker and slots. The agreement between the two big companies will make them the prime candidate to supply online casino games to Atlantic City’s Borgata Hotel which is owned by MGM. Certainly, the market conditions are in favor of Bwin.Party. With this, its investors are all geared up to purchase stock from the company and have reinstated their buy recommendation on the group. Daniel Stewart investors’ Michael Campbell predicts that New Jersey can turn up to be a $450 million market. This is especially true if the Internet sports betting is approved and offered intrastate. Securing a position within this market could drive significant profits to prime online casino operators and technology providers. He further explained that the market should demand a high multiple for fully regulated and taxed earnings. Because NJ has now secured its online gambling regulation, the group is reinstating its buy recommendation. Another online casino prime company 888.com also sees the state of NJ as a great venue for business expansion. According to the Morgan Stanley global finances firm, the UK based operator had its shares soar higher from 140p to 175p. Vaughn Lewis, who is an analyst from Morgan Stanley, says that casino online firms like Bwin.Party and 888 are the real big winners with a newly regulated and legalized online gambling market in New Jersey. It also led to the increase in their 2015 earnings per share forecast by 5 percent for Bwin.Party and 24 percent for 888. Lewis further explained that there will be more European operators who are going to jump in the online gambling bandwagon in New Jersey. William Hill to Buy Online Casino Software Developer Playtech William Hill, one of the world’s prime casino online providers, is projected to reach a market value of £5 billion in a span of five years. This is according to the company’s chief executive after he announced that it would be paying Playtech £424 million as its share of joint venture. Ralph Topping described the firm’s plan to buy 29 percent of stakes at William Hill Online as a major milestone. The plan of them getting into the online casino industry in the U.S. brightened even more when the New Jersey legislators approved the bill to permit the Internet and casino operations. Topping further explained that there was a strong support coming from shareholders for the deal with Playtech. It will be funded for a two to nine right issue at 245p thereby raising about £375 million. According to William Hill, the deal represents about 9.3 times the adjusted earnings before the interest, depreciation, tax and amortization. On the other hand, casino online prime games developer Playtech calculates it to reach about 11.5 times. Although the deal could be beneficial to both parties, there are some troubles that have to be handled first, including the staff walkouts and the injunctions. In fact, Topping once blogged saying that it is easier to get hold of Lord Lucan than his Playtech counterpart chief exec Mor Weizer. The doubts they had regarding each other had led to the involvement of a third party arbitrator that would help them find a perfect equation to the valuing of the 29 percent stake. However, it appeared that independent assessment is unnecessary simply because each party’s valuation was close to each other. Topping reiterated that both companies must win fairly. William Hill’s online revenues rose by 27 percent as reflected in its 2012 results. It equates to an overall increase of 12 percent. In only about seven weeks, the shares again rose by 29 percent. Topping was firm with his prediction that its prime online casino operations will reach £5 billion in five years. He also believed that there would be growth opportunities in the United States. Similarly, Playtech is eyeing the U.S. online gambling market while it continues to supply services to William Hill. Weizer also raised the possibility that Playtech would be using the proceeds that William Hill payout to acquire an online casino operator. However, it can’t be concluded for now. Record Breaking Tourney Comes Back in March March has just started but there are already torrents of offerings from prime casino online providers and poker rooms like Poker Stars. The Isle of Man based PokerStars.eu has recently released its schedule for its fourth Micro Millions festival. This promotion boasts 100 events from March 14th until the 24th and guaranteed prize pool of $5 million. Aside from the Micro Millions, Poker Stars will also launch one of the most awaited tournaments called Zoom Poker. There will be two events to be featured taking place on March 14 and 17 with $75,000 and $50,000, respectively. Zoom Poker only started a year ago, but it has already gained tremendous support from poker enthusiasts. This March, Zoom Poker will be open for tournament play. Players can still expect the Fast Fold option that allows them to move to another table with a shared pool of participants. The last three tables on the finals take participants to the standard fashioned tourney. The online casino prime provider seriously never stops giving out huge prizes to its players through a number of tourneys offered almost every month. During the debut of Zoom Poker last year, Andre Coimbra or also known as acoimbra played at the 100 Micro Millions joined the tourney and plans on doing the same this year as a part of his $100 to $100,000 bankroll challenge. He revealed that he will be playing a lot of Micro Millions tourneys in the coming weeks. What is great about this tourney is that players of all levels can join and get a chance to sit next to the prominent participants and win. On March 24, Micro Millions will be holding the $22 Main Event where the guaranteed prize is $1 million. On the opening day, there will be a $1 3x Tubro re-buy tourney with $150,000 guarantee. There will also be Sunday Storm specials that will be giving out a guaranteed prize of $300,000. The Micro Millions prepares about ten tourneys a day to cater for players of all levels, skills and preferred formats. Players all over the globe can participate in the tourneys because the casino online prime poker room Poker Stars makes sure that the schedule suits them all. Micro Millions is hailed as the most popular tourney series ever since Poker Stars launched it. A record breaking 1,308,994 entries were achieved last November. BetVictor Offers SNG Challenge with €50,000 Prize Pool Online casino prime provider BetVictor has recently launched its Sit N Go Challenge. This promotion kicks off on March 1 where players are given eight chances to win a part of the €50,000 prize pool. What players have to do is to complete the challenges and be on the top of one or more of the SNG Challenge leaderboards. Of course, they have to get the highest score in order to do so. Players will need to play through a block of consecutive qualifying SNGs. To get started, players will have to log in on the poker room software where they have to get into their account, and join the SNG Challenge found on the My Account section in the lower left corner. Players will then be directed to the SNG Challenge website. The challenge offers four buy in levels. Players can choose “less than €5,” “€5 to €9.99,” €10 to €29.99” and “€30+”. When players join the challenge, their nicknames will appear on the leader board that represents the stakes they opt to participate in. They need to complete a block of ten qualifying challenges under the same stake in order to get their very first leader board score. Players who are able to complete a third block of 10 qualifying SNGs and earned more points than the scores they got from the previous games will receive improved points on the leader boards. Players who complete the first block of 50 qualifying SNGs will see their first score on the Block of 50 leader board. This online casino prime offering features the all Euro six seat and nine seat Texas Hold ’em tables except the feeder, satellite, double up ad heads up tables. Each of the eight leader boards offers a total prize of €6,250. According to the prime casino online company head of poker, Andy Horne, the SNG Challenge is specially designed to test players’ skills over a range of single table tournaments. Another exciting thing to look forward to in the challenge is the €50,000 that is up for grabs. All players have to do is to qualify by playing through a minimum of 10 SNG challenges. Anyone can join and test their skills and luck. There are eight leader boards waiting for the eight best players. Mississippi Loses Potential Revenues with Legalization Rejection Online casino legalization prime supporter Mississippi Democrat representative Bobby Moak had again introduced the bill just recently for the second time at the House Ways and Means Committee. The bill that could have legalized operations in the state for poker and casino online and established the Mississippi Lawful Internet Gaming Act of 2013 has again been rejected. This has been the hot topic for many Internet gambling forums and on sites like vegasonlinecasino.cc. Online gamblers from in and out of state were very disappointed with the announcement posted on the Mississippi legislature’s website saying that the Bill HB 254 that is identical to its predecessor HB 1373 was also trashed as soon as it was introduced to the committee. The HB 254 could have brought Internet gambling to life in the state of Mississippi. It could have allowed the state to impose the five percent tax to prime casino online operators’ gross revenues. A huge part of the collected tax could have gone to the Mississippi Gaming Commission Fund that would run programs aimed to save problem gamblers and fight against Internet gambling related crimes. The HB 254 allotted 75 percent of these taxes for these programs while the rest will go to the General State Fund. It is sad to know though that it is no longer going to happen. The proposed bill also requires all license applicants to pay a non refundable deposit of $100,000 in order to be allowed to operate in the state. The proposed bill says that the operator would have to pay $100,000 and higher to the State General Fund every year and another $100,000 to the Mississippi Gaming Commission Fund. The renewal fees could have been set around $100,000. Prime online casino proponent Moak explained that his reason for re-introduction of the bill is to curtail illegal gambling operations. He said that the licensing system will protect the minors from these activities and fund programs for vulnerable individuals. The bill also included means and ways to collect more revenues from taxes, generate more jobs and finance developmental projects. Moak reiterate that if the bill was passed, there would have to be a strict criteria brought in place that would ensure that the people who will facilitate the industry and all casino online operators are of good character and fitness. Bally Technologies Prepares for Legalization Bally Technologies has very attractive company profile as presented in . The company is known globally for its specialty in the field of gaming systems, devices and technology solutions. It designs, operates, manufactures and distributes slots machines, lottery, video slots and central determination platforms and games. Even before the prime online casino sites came to life, Bally Technologies was already up and running with its history that can be traced back to 1932. However, there has been an interesting issue going around the industry today. This is pertaining to the huge amount of money that the chief executive officer is going to spend. Now, many are asking what lies ahead for the company. The recent report says that CEO Ramesh Srinivisian purchased 25,000 BYI shares at $47.48 for a total of $1,186,900. With this, company shares have gone up by $49.188 only a week later. According to reports, the reason behind the purchase is that the company is preparing for the upcoming legalization of casinos online in the United States. This is after the industry has learned that the Department of Justice announced that it is legal to gamble at prime online casino sites except in the case of sports betting. In September, Delaware already has its regulations in place. It will be followed then by Nevada, New Jersey and perhaps Massachusetts, Pennsylvania and possibly Hawaii. All states are gearing up for the possible launch of the for their residents. The possibility of having an interstate compact to offer these products and services is also being entertained. It appears that the Bally Technologies’ CEO is also seeing its future in the Internet gambling world. The company has already rooted out in the casino industry. Now, it seems like it is about to branch out to another venture targeting the United States. Ropes and Gray law firm representative David O. Stewart commented that there are millions of Americans who spend as much as $4 billion every year to their prime casino online games although there have been some opposition from the government. The once draconian laws in the US had driven more responsible offshore online casino providers from the market. With this, the US government has left nothing but unregulated operators. Fortunately, everything seems to have turned upside down now. As soon as it has secured its position, Bally Technologies is ready to plunge. Report Says Lotto is Another Culprit has again something to watch out for as more and more problem gamblers will be seeking for their help. This is mainly because it is not only the online casino prime offerings that are linked to the rise of problem gambling. The world’s favorite Lotto is also found to be one of the culprits. This is according to the consultation document released by The Ministry of Health’s Preventing and Minimizing Gambling Harm. Although casino online and offline still top the list, lottery lands next. According to the report, there has been an increase in gambling problems since Lotto started to offer massive jackpots four years ago. Many people see Lotto as simple and affordable game of chance that comes with a very low risk of turning someone into gambling addict. Unfortunately, this is not how the organization sees it. The report says that there have been 12 percent of clients that claim the Lotteries Commission as their main problem gambling outlet. Three years ago, it was only about 8.8 percent and rose to 11.5 percent in 2009 to 2010. The report pointed out that the increase is due to the frequently increasing large jackpot prizes and its enhanced appeal to people particularly in these times of poor economic growth. It further explained that the expansion of the Lotteries Commission offerings online has even boosted its popularity because of its convenience and accessibility in the same way that people experience with playing at prime online casino sites. One of the biggest jackpot prizes was won in 2009 with as much as $27 million. Just about a year after, another jackpot prize was again given out with a huge $28.7 million win. A year ago, the second highest jackpot prize was revealed with a whopping $30 million. It was mainly due to the rising sales in tickets. However, there were two winners at that time. As a result, they divided the prize equally. Additionally, the report suggests that there has been a dramatic increase in the number of individuals that claim Lotto as their main gambling outlet. From 6.5 percent in 2009 to 2009, it reached 7.3 percent from 2010 to 2011. The report does make sense though because Lotto is actually a lot cheaper than playing at casino online prime offerings. However, further studies and research should be done to support this claim. Jailbroken Devices Not Suitable for Mobile Casinos iPhone, iPad, smartphones and new generation mobile devices have become the prime online casino device today. This is because it takes the gambling world right at the palm of your hands. With these gadgets on hand, anyone can play at sites like freeonline-slots.com wherever and whenever. However, reports say that there are about 7 million Apple users that use jailbreaking software such as Cydia to bypass the App store. Many mobile gamers may not see the repercussions of doing so, but Apple has already warned its users to avoid such methods. This is because it does not only pose harm to the device but also to the users. Jailbreaking of Apple devices is the process of using a hardware or software to root access to the iOS operating system. This method allows the owners to download applications and extensions that are not available in the Apple’s official App store. Just recently, professional jailbreakers were able to apply the same method to the updated iOS 6 operation system. Apple highly discourages these operations. Unfortunately, millions have already using it including some casino online players. Apple explained that the iOS has been designed to be more reliable and secure. Its prime features can protect the device against viruses and malware. It also protects all personal information that is being stored in there. Online casino prime players should do away with this upgrade. It is important that they are extremely careful with all the apps they are using. This is especially true when it is a third party app. This is mainly because the jailbroken devices that are downloaded with apps that come from unsecure sites make the device more prone to being hacked. Most mobile casino gamers keep their personal data on their phone including their bank accounts and passwords to their casino online account. All these can be accessed easily by hackers without their knowing. It is always best to choose a reputable mobile app and site to play casino at. These apps should be accessible to users without having to do the jailbreaking. Today, there are already a lot of mobile casino apps from trustworthy gaming provides such as the Jackpot City Mobile, All Slots, the Royal Vegas and many others.

1. Field of the Invention The present invention relates to dental care facilities and in particular to a central vacuum system for dental care operatories incorporating an integrated amalgam separation system. 2. Description of the Prior Art Modern dental facilities usually include multiple operatories and a central vacuum system. Dental aspirator tips are provided at each operatory for disposition in the patient""s oral cavity to remove aerosols, liquids, solid debris and odors from the patient""s mouth. Typical conventional dental vacuum systems have been far from ideal from the standpoints of noise output, vacuum intensity and flow rate characteristics, efficiency, reliability and recovery of amalgam metals, and particulate matter used or generated in the dental process. Additionally, environmental laws at some locations now have incorporated mandatory amalgam separation in the dental suction system. The dental vacuum system of the present invention as described herein incorporates an air/liquid separation tank having an integral sedimentation amalgam separator and collector insuring capatability of the system with the current state of the law. An object of the present invention is to provide for a novel dental vacuum system for the efficient evacuation of aerosols, liquids, solid debris and odors from the patient""s mouth in a dental operatory. A still further object of the present invention is to provide for a novel dental vacuum system having a novel air/liquid/solids separation tank which incorporates an integral amalgam separation and collection capability which is removably replaceable from the system. A dental vacuum system for use in dental operatories in which a dental aspirator tip in the dental operatory is in fluid communication with a vacuum means for providing suction to the system, there being interposed between the dental aspirator tip and the vacuum means, a separation chamber for the separation of effluent and solids including amalgam from the air stream. Solids including amalgam and liquids are collected in the separation chamber while air is passed through for eventual venting to the atmosphere. When no vacuum is required in the operatories, a sensor turns off the vacuum means and a timer relay allows for sedimentation to occur in the chamber. Thereafter, contaminated water and solids including amalgam are drawn out to an amalgam separation and collection chamber. Contaminated water is drawn off to the drain and the process repeats itself until the amalgam collection chamber has accumulated its maximum amount of solids, including amalgam. It is then prepped for recycle and replaced with an empty amalgam collection chamber and the full chamber is sent out for recovery of precious metals and proper disposal of hazardous material.

Retained medullary cord extending to a sacral subcutaneous meningocele. A retained medullary cord (RMC) is a rare closed spinal dysraphism with a robust elongated neural structure continuous from the conus and extending to the dural cul-de-sac. One case extending down to the base of a subcutaneous meningocele at the sacral level has been reported. We report on three cases of closed spinal dysraphism, in which a spinal cord-like tethering structure extended out from the dural cul-de-sac and terminated at a skin-covered meningocele sac in the sacrococcygeal region, which was well delineated in curvilinear coronal reconstructed images of 3D-heavily T2-weighted images (3D-hT2WI). Intraoperative neurophysiology revealed the spinal cord-like tethering structure was nonfunctional, and histopathology showed that it consisted of central nervous system tissue, consistent with RMC. The tethering structure histologically contained a glioneuronal core with an ependymal-like lumen and smooth muscle, which may indicate developmental failure during secondary neurulation. When the RMC extending to a meningocele is demonstrated with the detailed magnet resonance imaging including 3D-hT2WI, decision to cut the cord-like structure for untethering of the nervous tissue should be made under careful intraoperative neurophysiological monitoring.

--TEST-- XMLReader: accessing empty and non existing attributes --SKIPIF-- <?php if (!extension_loaded("xmlreader")) print "skip"; ?> --FILE-- <?php /* $Id$ */ $xmlstring =<<<EOF <?xml version="1.0" encoding="UTF-8"?> <foo bar=""/> EOF; $reader = new XMLReader(); $reader->XML($xmlstring); $reader->read(); var_dump($reader->getAttribute('bar')); var_dump($reader->getAttribute('baz')); $reader->close(); $xmlstring =<<<EOF <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE foo SYSTEM "012.dtd"> <foo bar=""/> EOF; $xmlstring = str_replace('012.dtd', dirname(__FILE__).'/012.dtd', $xmlstring); if (DIRECTORY_SEPARATOR == '\\') { $xmlstring = str_replace('\\',"/", $xmlstring); } $reader = new XMLReader(); $reader->XML($xmlstring); $reader->setParserProperty(XMLReader::DEFAULTATTRS, true); while($reader->read() && $reader->nodeType != XMLReader::ELEMENT); var_dump($reader->getAttribute('bar')); var_dump($reader->getAttribute('baz')); $reader->close(); echo "\nUsing URI:\n"; $reader = new XMLReader(); $file = dirname(__FILE__) . '/012.xml'; if (DIRECTORY_SEPARATOR == '\\') { $file = str_replace('\\',"/", $file); } $reader->open($file); //$reader->setParserProperty(XMLReader::DEFAULTATTRS, true); while($reader->read() && $reader->nodeType != XMLReader::ELEMENT); var_dump($reader->getAttribute('bar')); var_dump($reader->getAttribute('baz')); $reader->close(); $reader = new XMLReader(); $reader->open(dirname(__FILE__) . '/012.xml'); $reader->setParserProperty(XMLReader::DEFAULTATTRS, true); while($reader->read() && $reader->nodeType != XMLReader::ELEMENT); var_dump($reader->getAttribute('bar')); var_dump($reader->getAttribute('baz')); $reader->close(); ?> ===DONE=== --EXPECT EXACT-- string(0) "" NULL string(0) "" string(0) "" Using URI: string(0) "" NULL string(0) "" string(0) "" ===DONE===

University College of the North University College of the North (UCN) is a post-secondary institution located in Northern Manitoba, Canada. UCN has a student body of approximately 2,400 annually, and a staff of approximately 400. The Chancellor of UCN is Edwin Jebb. History Northern Manitoba Vocational Center was established in The Pas in 1966 and the Thompson campus was opened in the early 1980s. University College of the North was established on July 1, 2004, with the passage of the University of the North Act in the Legislative Assembly of Manitoba, and is the successor to Keewatin Community College. Campus UCN has two main campuses, in The Pas and Thompson. There are 12 regional centres in Churchill, Cross Lake, Easterville, Flin Flon, Pukatawagan(Mathias Colomb), Grand Rapids (Misipawistik), Nelson House (Nisichawayasihk), Norway House, Oxford House (Bunibonibee), St. Theresa Point, Split Lake, and Swan River. Nine of UCN's 12 regional centres are in First Nations communities. In partnership with local Cree Nation authorities, UCN co-sponsors public libraries in Pukatawagan (Mathias Colomb), Norway House and Chemawawin (Easterville). Programs UCN offers more than 40 degree, diploma, and certificate programs in the Faculty of Arts and Science, Faculty of Trades and Technology, Faculty of Health, Faculty of Education, Faculty of Business, and as part of community-based contract training and Apprenticeship training. Aboriginal The UCN Council of Elders provides guidance through the sharing of traditional knowledge, beliefs and values. There are Aboriginal Centres at UCN's two main campuses. The Mamawechetotan Centre in The Pas and ininiwi kiskinwamakewin Centre in Thompson offer programs that promote cross-cultural awareness. There is a great deal of diversity that is well respected. Scholarships and bursaries The Government of Canada sponsors an Aboriginal Bursaries Search Tool that lists over 680 scholarships, bursaries, and other incentives offered by governments, universities, and industry to support Aboriginal post-secondary participation. University College of the North scholarships for Aboriginal, First Nations and Métis students include: Manitoba Hydro Second Year to Final Year Engineering Technology Bursary See also Education in Canada List of universities in the Canadian Prairies Higher education in Manitoba References External links University College of the North (Official Site) University College of the North (Canadian Universities) Category:Universities and colleges in Manitoba Category:Educational institutions established in 2004 Category:The Pas Category:Education in Thompson, Manitoba Category:2004 establishments in Manitoba Category:Higher education in Canada

As Reported by David Hultsman of EOTT: We began the meeting with a review of the progress to date. Detail transaction flows were discussed including the very important items, Price and Volume. Both will be required daily in groupings. The changes by day in the positions was stated as already being in place. We then began to discuss what we needed to have and what was already present in the current system. Items such as Groups, Price, Priority, Today's Position, Change in Trades, etc were all mentioned as required. The following is my version of the meeting yesterday. I hope I captured everything correctly and have the proper "to do" list included: We decided that we needed to provide contract details on Evergreen contracts and that a copy of a delivery contract needs to be provided to the team. A matrix was created which defined which barrel information was in the current system and which information was needed but not available. POSITION REPORT STATUS CURRENT BARREL INFORMATION IN THE REPORT INFORMATION NOT IN THE REPORT Exchange Traded barrels Evergreen Broker barrels (exchange contracts) Term to Evergreen Spot barrels (one day lag in entry) Lease Term barrels Pipeline barrels OTC Storage barrels Swaps Inventory with a sale against them are Options the only barrels in the report As additional tasks we stated that Tickets to Actuals need to be in the Report everyday. To achieve this we need Right Angle ticket detail actualized daily. We then stated that we need to develop a plan to achieve the above stated needs. We then agreed on the following: 1. Create the step by step plan to achieve the daily information goals. 2. We will define the information fields needed. Lori and Yolanda will work with the DataBase Manager to achieve the necessary information. 3. We will create the plan to build the analytical data mart from which this information will be extracted and reported. 4. The team will consist of Bjorn Hagelmann, David Hultsman, Lori Maddox, Lawrence Clayton, Sheila Harris, Biaggio McPhee, Lisa Meyer, Jim Martel. alternates may be decided at a future date. We will meet each Wednesday at 730 am in conference room 5830.

[ { "description": "maximum validation", "schema": {"maximum": 3.0}, "tests": [ { "description": "below the maximum is valid", "data": 2.6, "valid": true }, { "description": "above the maximum is invalid", "data": 3.5, "valid": false }, { "description": "ignores non-numbers", "data": "x", "valid": true } ] }, { "description": "exclusiveMaximum validation", "schema": { "maximum": 3.0, "exclusiveMaximum": true }, "tests": [ { "description": "below the maximum is still valid", "data": 2.2, "valid": true }, { "description": "boundary point is invalid", "data": 3.0, "valid": false } ] } ]

Howard E. Babcock Howard Edward Babcock (February 23, 1889 – July 12, 1950) was the Chairman of Cornell Board of Trustees from 1940 to 1947. He served on the Federal Farm Board as well as on the board of the Central Bank for Cooperatives. He was co-president of the National Cooperative Council and chairman of the American Institute of Cooperation. Biography He was born on February 23, 1889. He graduated from Syracuse University and studied agriculture in the Cornell Summer School of 1911. He then taught school and later became a professor of marketing in the Cornell University College of Agriculture and Life Sciences. While he was a professor, the Cooperative Grange League Federation Exchange, a failing farmer's cooperative, requested Babcock's help. Under his management, the cooperative became the largest farmer's cooperative in the world. Upon becoming the President of the New York State Grange, Babcock became an ex officio Cornell Trustee in 1930. He served as Chairman of the Board from 1940 to 1947. He died on July 12, 1950. Legacy He was inducted into the Cooperative Hall of Fame in 1976. External links NY State Grange References Category:Syracuse University alumni Category:Cooperative organizers Category:1950 deaths Category:1889 births

Antisense SNF1-related (SnRK1) protein kinase gene represses transient activity of an alpha-amylase (alpha-Amy2) gene promoter in cultured wheat embryos. A DNA fragment corresponding to part of an SNF1 (sucrose non-fermenting-1)-related protein kinase (SnRK1) transcript was amplified by a polymerase chain reaction (PCR) from a wheat (Triticum aestivum) endosperm cDNA library. It was used to construct a chimaeric gene, pUasSnRKN, comprising a ubiquitin promoter, the SnRK1 PCR product in the antisense orientation and the nopaline synthase (Nos) gene terminator. This construct was used in transient gene expression experiments in cultured wheat embryos together with a series of reporter gene constructs. These included the wheat alpha amylase gene alpha-Amy2 promoter with UidA (Gus) coding region (palpha2GT), rice actin promoter with Gus (pActIDGus), ubiquitin promoter with Gus (pAHC25) and actin promoter with green fluorescent protein (GFP) gene (pAct1Is-GFP1). All of the reporter genes were found to be active when bombarded into scutellae isolated from immature grains at 25 d post-anthesis and incubated on MS medium for 24 h prior to bombardment. However, co-bombardment of palpha2GT with equimolar amounts of pUasSnRKN resulted in no detectable Gus activity, indicating that the antisense SnRK1 construct repressed the alpha-Amy2 promoter. Co-bombardment with pUasSnRKN had no effect on the activity of the other promoters used in the study. A triple bombardment with palpha2GT, pAct1Is-GFP-1 and pUasSnRKN resulted in clear green fluorescence, indicating that the bombarded cells were still viable, but no Gus activity. RT-PCR analysis showed clearly that the antisense SnRK1 gene was expressing. Northern and RT-PCR analyses confirmed that SnRK1 and both alpha-amylase genes, alpha-Amy1 and alpha-Amy2, are expressed in cultured wheat embryos harvested from grain 25 d post-anthesis. Expression of alpha-Amy1 and alpha-Amy2 was up-regulated by sugar starvation.

'You're the best I have,' Panetta tells USS Stennis sailors By Adam Ashton The News Tribune Published: August 23, 2012 More than 2,000 servicemembers stand in formation aboard the USS John C. Stennis as Defense Secretary Leon Panetta walks to the stage on August 22, 2012, where he thanked them for their efforts in preparing for their upcoming deployment to the Middle East. Steve Ringman/Seattle Times/MCT Related Leon Panetta used his first visit to the Puget Sound area as U.S. defense secretary to thank a Bremerton-based aircraft carrier crew that accelerated its deployment schedule because of Middle East unrest. He also took the opportunity to foreshadow budget decisions that could reshape the military in Washington state. Panetta is sending the 5,500 sailors in the USS John C. Stennis carrier group back to the Persian Gulf four months ahead of schedule to meet threats in nations such as Syria and Iran. The Stennis is due to leave port on Monday. "You're the best I have," Panetta told about 2,200 sailors who assembled for an "all-hands call" on the Stennis. The ship returned from the Persian Gulf in March and anticipated another mission at the end of this year. Panetta in July expedited the deployment because of the uncertain security situation across the Middle East. The ship will play a role in monitoring Syria's apparent civil war and in protecting shipping lanes from Iranian threats in the Strait of Hormuz. The Puget Sound area is a major hub for the nation's military, but public visits from the Pentagon's top civilian are rare. Former Defense Secretary Donald Rumsfeld visited Fort Lewis in 2002; his successor, Robert Gates, came to the post in 2008. Panetta did not go to Joint Base Lewis-McChord on Wednesday. He said he was up to date on ongoing controversies about behavioral health diagnoses at Lewis-McChord's Madigan Army Medical Center. Advertisement The Army has reinstated Madigan's commander after a six-month hiatus and eliminated the widespread use of forensic psychiatrists who checked and sometimes changed post-traumatic stress diagnoses among war veterans. Panetta asked each service branch to review and standardize its procedures for PTSD diagnoses. "We've made some changes," he said Wednesday. "I'm awaiting further recommendations as to what we can do to do everything possible to make sure that that kind of misdiagnosis doesn't take place." He outlined to sailors his priorities for maintaining the military's strength even as the nation prepares to cut defense spending. He said he aims to: --Sustain promised benefits to current service members. --Provide training and incentives so the force does not become "hollow." --Turn the Pentagon's attention to challenges in the Pacific region. --Invest in cyber warfare and technology. --Develop rotations for military units to train with and cultivate foreign allies. --Ensure U.S. forces can defeat more than one enemy at once. In remarks to reporters, he said the Puget Sound defense community is positioned to thrive as the Pentagon turns to the Pacific. "I would assume that Bremerton's going to stay busy for a long time to come as a result of having to focus and project our force into the Pacific," he said. "We're going to need the kind of shipyards and maintenance and backup that Bremerton provides." Panetta said he understood he was asking a lot of sailors who only recently returned from seven months at sea. "I know it can be tough, but it's also something that makes us proud," knowing that the sailors are ready to deploy, he said. Many sailors had their minds on their last days at home before leaving on another long mission. They cheered when they learned most would have Friday off. "Deployments are fun. They say it's like a big, floating city, and it really is," said Petty Officer 2nd Class Vanetta Randle, 24, of Silverdale. One sailor characterized the past two years of life on the Stennis as unusually busy. Panetta said the ship has been at sea about 70 percent of the time over that period. "It's tough," said Petty Officer 1st Class Ricky McCaig, 29, of Belfair. Frequent deployments "can harm both sides with family and service members; both can lose morale." He and other sailors said they're accustomed to changing plans at a moment's notice. McCaig said he appreciated seeing Panetta just before the next assignment. "We're trained to just follow orders and go where we're told," he said. "It's definitely a privilege to have someone doing all the telling come to talk to us about where we're going." Naval Base kitsap project gets OK The Army Corps of Engineers on Wednesday cleared the way for the largest defense construction project in Washington state -- a $715 million effort to improve the wharf that serves eight nuclear submarines at Naval Base Kitsap. The corps issued a final construction permit for the project. The Navy plans to break ground next month, a spokeswoman said. The Defense Department's 2012 budget included $280 million to begin the work, and the Navy gave its final approval in May. Washington lawmakers including Rep. Norm Dicks, D-Belfair, have endorsed it as necessary to sustain the base's eight Trident ballistic-missile submarines. Anti-nuclear advocacy groups in June sued to block the project, arguing that the Navy should have considered other alternatives.

Q: Equivalence between categories of affine schemes over $X$ and representable functors $\operatorname{Points}(x) \to \operatorname{Sets}$ I am reading Strickland - Formal schemes and formal groups. For a functor $X\colon \operatorname{Rings}\to \operatorname{Sets}$, he defines (2.14) the category of $\operatorname{Points}(X)$ in the following way: Objects: $(R,x)$ with $x \in X(R)$. Morphisms: ring morphisms $f\colon R\to S$ s.t $X(f)(x)=y$. In remark 2.15 he claims that the category $\mathcal{X}_X$ of affine schemes $Y$ over an affine scheme $X$ and the category of representable functors $Y'\colon \operatorname{Points}(X)\to \operatorname{Sets}$ are equivalent using $Y(S)= \coprod_{z \in X(S)} Y'(S,z)$. I can't really understand how the equivalence follows. Would appreciate any help. P.S. I understood the equivalence of (a) and (c)—this is usually referred to as the description of a scheme using its functor of points. A: You want to see that the category $\mathcal{X}_X$ of affine schemes over $X$ (or affine $X$-schemes) is equivalent to the category of representable functors $Y':\text{Points}(X)\to \text{Set}$. First, given an affine $X$-scheme $f:Y\to X$, and a point $(x,R)$ with $x\in X(R)$ and $R$ a ring, we associate the set $$f(R)^{-1}(\{x\})=\{ y\in Y(R) \ | \ f(R)(y)=x\},$$ where $f(R):Y(R)\to X(R)$ is the associated map on the $R$-points (usually called just $f$ by abuse of notation). So the functor say $f'$ (bad notation, by sure) we associate is on the objects of $\text{Points}(X)$ defined by $$f'((R,x)):=f(R)^{-1}(\{x\}).$$ (Note that an $X$-scheme is better defined by the morphism $f$ that by the scheme $Y$). On the contrary, that a functor $Y':\text{Points}(X)\to \text{Set}$ is representable it means (my guess) that there exists a ring $S$ and a point $y\in X(S)$ such that $$Y'(R,x)=\{h:R\to S \ | \ X(h)(x)=y\}.$$ Note that a point $y\in X(S)$ is the same that a map $y:\text{Spec}(S)\to X$. We associate then to $Y'$ such affine $X$-scheme $y$.

Identification and characterization of 12 microsatellite loci in Cnidium dubium (Apiaceae) using next-generation sequencing. We present a set of 12 polymorphic nuclear microsatellite loci, identified for the first time within the herbaceous plant species Cnidium dubium (Schkuhr) Thell. using next-generation sequencing. To characterize these loci, 40 plants were analyzed genetically. The number of alleles ranged from 2 to 23, with an average of 8.3 alleles per locus. Mean observed and expected heterozygosities ranged from 0.25 to 0.95 and 0.30 to 0.94, respectively. All 12 nuclear microsatellite loci showed scorable and polymorphic fragments after polymerase chain reaction. The new marker set can be used for future studies of genetic diversity and differentiation as well as estimation of gene flow and spatial genetic structures.

Segambut Segambut is a sub-district and a parliamentary constituency in Kuala Lumpur, Malaysia. From the high-end condominiums of Mont Kiara and Sri Kiara to the middle-class areas of Taman Sri Segambut and Bandar Manjalara, and the rural areas of Segambut Dalam and Kampung Sungai Penchala, the Segambut constituency has a diverse population base. Segambut once belonged to the mukim of Batu Caves before 1974, when Kuala Lumpur was still part of Selangor. Segambut is served by the train station of the same name, on the KTM Komuter line, that runs from Tanjong Malim to Port Klang, via KL Sentral and other city-centre stations. The new Istana Negara is located in this constituency, along Jalan Tuanku Abdul Halim (fka Jalan Duta). Townships Kampung Pasir Segambut Segambut Dalam Segambut Luar Segambut Tengah Segambut Bahagia Segambut Damai Segambut Permai Segambut Tambahan Taman Sri Segambut Taman Segambut Segambut Garden Bukit Segambut Taman Bukit Maluri Kampung Sungai Udang Kampung Concrene Taman Kok Doh Taman City Kanan Taman Sri Sinar Kampung Kasipillay Taman Bunga Kampung Sungai Pencala Medan Damansara Bukit Kiara Sri Hartamas Taman Tun Dr Ismail (TTDI) Utara TTDI Selatan Bandar Manjalara Bukit Damansara Education The French School of Kuala Lumpur is located in Segambut. Its current campus opened in 2005. Politics The constituency was created in 1994 and has been considered to be a Barisan Nasional safe seat until the 2008 General Elections. Datuk Dr Tan Kee Kwong of Parti Gerakan Rakyat Malaysia (Gerakan) had won the seat the last three terms. However, in the 2008 General Elections, Lim Lip Eng from the Democratic Action Party (DAP) won the seat. The current MP for Segambut is Hannah Yeoh Tseow Suan from Pakatan Harapan-DAP which is former ADUN . Districts Segambut is one of Kuala Lumpur's eleven districts serve as administrative subdivisions under the Kuala Lumpur City Hall authority. References Tan, Karr Wei. "The two faces of Segambut" (Archive). The Star, 23 February 2008. Category:Suburbs in Kuala Lumpur

Q: Select option won't 'trigger' on the first value I have this working snippet which console log the selected value from the dropdown. The problem is, if I select the first option (Missing Promo, which is already selected), it won't 'trigger' the onChangeProblem function. How can I resolve this? Thank you for your time! A: That's because the first option is selected by default but hasn't changed. change event will only get triggered if there is a change in the selected option. You should change your template like this: <select (change)="onChangeProblem($event)"> <option disabled value="null">Select an Option</option> <option *ngFor="let problemStatement of problemStatementArray" [value]="problemStatement"> {{problemStatement}} </option> </select> Since there would be a placeholder text, one an option is selected, the change event will trigger, making your change handler to run. That's how it's ideally implemented. Here's a Sample StackBlitz for your ref.

Multitarget stepping program in combination with a standardized multicomponent exercise program can prevent falls in community-dwelling older adults: a randomized, controlled trial. To investigate the effects of a twice-weekly multitarget stepping (MTS) task combined with a multicomponent exercise program on stepping accuracy, gaze behavior, fall risk factors, and fall rates. Randomized controlled trial. Community. Community-dwelling adults aged 65 and older in Japan (N = 264) were randomized into an MTS group (n = 132) and a control group (n = 132). Twenty-four weeks of a twice-weekly community-based MTS program in combination with a standardized multicomponent exercise program. Number of falls and fall-related fractures during a 12-month follow-up period after completion of the intervention, stepping accuracy, gaze behavior while performing the MTS test, and results of four clinical tests relevant to assessment of risk of falls (Timed Up and Go (TUG), functional reach (FR), 10-m walking, and Five Chair Stand (5CS)) were measured. Thirteen participants (11.6%) in the MTS group and 39 (33.0%) in the control group fell during the 12-month follow-up period (incidence rate ratio = 0.35, 95% confidence interval (CI) = 0.19-0.66). Three participants in the MTS group and 13 in the control group experienced fall-related fractures during the 12-month follow-up period (relative risk = 0.22, 95% CI = 0.06-0.80). After the intervention, a subset of participants in the MTS group had significantly greater improvement in stepping accuracy and gaze behavior during the MTS test, and all participants in the MTS group had significantly greater improvement in performing the TUG and 10-m walking (P < .001). Participants who performed MTS tests combined with a multicomponent exercise program showed greater improvements in stepping accuracy, gaze behavior, and physical performance in a virtually complex environment. Considering the less-frequent fall rate in the follow-up assessment, these improvements could contribute to preventing falls in community-dwelling older adults.

(ANTIMEDIA) United Kingdom — Over one billion people are active daily on Facebook, and as a result, the social networking giant is continually updating its products and services. The most recent attempt to expand resources available to users was rolled out last week in the U.K., with a feature that

Q: Segmentation fault при обращении к указателю Решил для развития кругозора написать собственную имплементацию аллокатора памяти под Linux, но при написании столкнулся со следующей проблемой. Вызов sbrk() возвращает указатель на начало выделенной памяти, затем я этот указатель привожу к указателю на структуру Hdr. Но когда я пытаюсь задать значение переменной по этому указателю, программа вылетает с Segmentation error. Чутьё подсказывает, что я делаю глупость, ведь структуру нужно инициализировать прежде чем использовать, но я не знаю как в данной ситуации правильно это сделать, учитывая, что память под структуру уже выделена вызовом sbrk(). Заранее благодарен за любую помощь! struct Hdr { int is_avaible; int size; }; void *malloc(long memory_size) { void *memory_location; struct Hdr *current_location_struct; ... memory_location = sbrk(memory_size); if (memory_location == NULL) return; current_location_struct = (struct Hdr *)memory_location; current_location_struct->is_available = 0; ... } A: Давайте посмотрим как работает sbrk(). Схематично: void *sbrk(intptr_t increment) { void *heap_ptr = current_heap_end; if( increment > 0 ) { if( !expand_process_heap( increment ) ) { heap_ptr = (void *)-1; } } else if( increment < 0 ) { heap_ptr = reduce_process_heap( -increment ); } return heap_ptr; } Рассмотрим с конца. Если increment равен 0, то возвращаем текущий адрес конца кучи. Если отрицателен, то уменьшаем кучу и возвращаем адрес нового её конца. И если он положителен, то увеличиваем кучу и возвращаем старое его значение. Вот как раз пункт 3 ошибочно многими (не только начинающими программистами, но и книгописателями!) воспринимается как некий способ выделить память. Формально - да, как бы "память выделяется", как раз размером increment байтов. Но на деле эту память использовать по своему усмотрению напрямую нельзя! Потому что пытаясь туда что-то записать вы "сбиваете настройки" системному менеджеру памяти, что закономерно приводит к краху программы. Конечно, в случае реальной программы, а не в учебных примерах, в которых после использования sbrk() и вывода каких-то демо-значений в программе больше ничего не происходит. P.S. Кстати, если бы вы перед чтением сомнительной :) литературы прочитали man sbrk, то обратили бы внимание на следующее: On success, sbrk() returns the previous program break. (If the break was increased, then this value is a pointer to the start of the newly allocated memory). On error, (void *) -1 is returned, and errno is set to ENOMEM. P.P.S. Я уж не говорю о том, что заменить библиотечный malloc() своей реализацией - это не просто выстрелить себе в ногу, это усесться на связку гранат, облиться бензином и подорвать их.

I've seen a lot of articles about retrieving the identity ID of a new created record using other DBs and frameworks. I am creating a record in code in a Silverlight RIA Services app. I perform the Add method and then do a SubmitChanges on the DomainContext, but I do not see the ID. It comes up as 0. How do I get the new ID? Thanks.

Before the Affordable Care Act was fully implemented and Indiana’s Medicaid expansion program, the Healthy Indiana Plan, the U.S. Census Bureau data showed 14 percent of Hoosiers without health insurance. Director of public policy for Covering Kids and Families Mark Fairchild says the program helped decrease that number. "Even by expanding late, Indiana was able to reduce its rate substantially compared to those that did not make that decision at all," says Fairchild. The most recent national data show the state holding at 8 percent without medical insurance. But Fairchild says thousands were dropped from HIP 2.0 this year, after enrollment changes – these are not recorded in the most recent Census report. The upcoming work requirement changes may also increase the number of uninsured. "We know consumer confusion is the easiest way to make people back away from a program, when they don’t know what they’re getting into and what some of the rules might be," says Fairchild. Fairchild says still, HIP 2.0 benefits from positive feedback. "One of the positives with the Healthy Indiana Plan is that people really do view it as a public good, they don’t put it in the same bucket with terms we don’t like – like welfare," says Fairchild.

Stop It. Google Won't Buy Twitter. - lotusleaf1987 http://battellemedia.com/archives/2010/09/stop_it_google_wont_buy_twitter ====== yhvh Could someone make all scientific journals free to read, I'm super fucked off stealing access through various student friends accounts. ~~~ yhvh I definitely value the fact that approving the content of journals is a costly business, it's just a frustration of mine. I apologise for the tone. ~~~ mayank It's a costly business for the people doing the actual reviewing, ie the scholars who essentially work for free. Academic publishing is, I hope, the next industry to be reformed by open access web archives, although I'll be the first to admit that it's a long way away. So I think it's perfectly fine for you to be "fucked off". ~~~ carbocation Despite accepting basically everything that is valid science (typically a much lower bar than is required to publish in most journals), PLOS One has had a consistently increasing impact factor. In other words, the open access approach is getting some respect from scientists who produce the original material. ------ sahaj google may eventually buy twitter, but they'll wait for the valuation to drop. twitter is an awesome technology, but i'm not sure how well and how long someone can monetize it. ~~~ alain94040 Delayed tweets could make money. [http://blog.fairsoftware.net/2010/09/30/a-real-business- mode...](http://blog.fairsoftware.net/2010/09/30/a-real-business-model-for- twitter/) ~~~ aberkowitz Failure of similar idea - Diggnation used to have a tiered system - subscribe and get everything early, don't subscribe and wait a day or two. This worked until someone thoughtfully [and illegally] created a mirror that provided subscriber time access to non subscribers. I believe that Diggnation no longer has a subscription service.

860 F.2d 1080 Unpublished DispositionNOTICE: Sixth Circuit Rule 24(c) states that citation of unpublished dispositions is disfavored except for establishing res judicata, estoppel, or the law of the case and requires service of copies of cited unpublished dispositions of the Sixth Circuit.UNITED STATES of America, Plaintiff-Appellee,v.Oscar FLORES, Defendant-Appellant. No. 88-1694. United States Court of Appeals, Sixth Circuit. Oct. 25, 1988. Before NATHANIEL R. JONES and ALAN E. NORRIS, Circuit Judges, and CONTIE, Senior Circuit Judge. ORDER 1 This matter is before the Court upon consideration of this Court's September 19, 1988 order directing the defendant to show cause why this appeal should not be dismissed for lack of jurisdiction. On December 29, 1987, the district court entered the judgment and commitment order sentencing the defendant on separate counts of distribution of heroin and possession of a firearm by a convicted felon. The defendant filed a notice of appeal with the district court on May 31, 1988, one hundred ten (110) days late. The defendant has responded, conceding the untimeliness of his appeal. 2 Rule 4(b), Fed.R.App.P., provides that a notice of appeal in a criminal action must be filed within ten days from the entry of judgment. The final judgment for purposes of 28 U.S.C. Sec. 1291 in a criminal action is the sentence. Berman v. United States, 302 U.S. 211, 212 (1937). Timely filing of a notice of appeal is a mandatory and jurisdictional prerequisite which this Court can neither waive nor extend. Browder v. Director, Dep't of Corrections of Illinois, 434 U.S. 257, 264 (1978). The appellant may move the district court for an extension of time in which to appeal upon a showing of excusable neglect, provided that such a motion is made within forty (40) days from the entry of judgment. Rule 4(b), Fed.R.App.P. In the present case, however, such an extension of time was neither sought nor granted. Accordingly, 3 It is ORDERED that the show cause order entered September 19, 1988, is hereby discharged and this action is dismissed for lack of jurisdiction. Rule 9(b),1 Rules of the Sixth Circuit.

The present invention relates to the utilization of magnets, and especially to the development of usable power from assemblies of permanent magnets, which exert a magnetic field of varying intensity, depending on the strength of the magnet. To the inventor's knowledge, permanent magnets have not been used in the manner set forth herein to develop usable power.

Q: Web Scraping using simplehtmldom on multiple sites I am using simplehtmldom parser for my recent web scraping project and the project is actually building a price comparing website build with CodeIgniter. The website has to fetch product names, description and price from different shopping websites. Here is my code: $this->dom->load_file('http://www.site1.com'); $price1 = $this->dom->find("span[itemprop=price]"); $this->dom->load_file('http://www.site2.com'); $price2 = $this->dom->find("div.price"); $this->dom->load_file('http://www.site3.com'); $price3 = $this->dom->find("div.priceBold"); $this->dom->load_file('http://www.site4.com'); $price4 = $this->dom->find("span.fntBlack"); $this->dom->load_file('http://www.site5.com'); $price5 = $this->dom->find("div.price"); The above code takes approximately 15-20 seconds to load the result into the screen. When I try with only one site, it just takes 2 seconds. This is how the simplehtmldom works with multiple domains? Or is there a way to optimize it? A: PHP Simple HTML DOM Parser has some memory leak issue, so before trying to load a new page, clear the previous one using: $this->dom->clear(); unset($this->dom); If this doesn't change anything, then one of your websites is taking much time to respond... you'll have to check one by one to find the culprit xD

Afshar Beylik Afshar Beylik was a Turkish beylik (principality) in Eastern Anatolia in the early 16th century.(Afşar and Avşar are alternative names) It was founded by the Afshar tribe which was an Oghuz Turk tribe with Kızılbaş faith. Emergence In 1473 Mehmet II of the Ottoman Empire defeated Uzun Hasan of Akkoyunlu Turcoman Empire which was controlling the Eastern Anatolia (as well as territories in Iraq, Iran and Azerbaijan) This defeat was a severe blow to Akkoyunlu prestige and they lost their control on the north territories of the empire (which is roughly the north part of the Eastern Anatolia.) During the choas, Sevindik Bey of Afshar tribe emerged as the new sovereign of the area in 1480s. The beylik At the peak of its power the beylik controlled the area between Alagyaz (now in Armenia) in the east and Kop Dağı (a mountain to the west of Erzurum, Turkey) in the west. When Akkoyunlu dynasty was replaced by Safavids (who were relatives of Akkoyunlu) Sevindik Bey formed a loose alliance with Ismail I of Safavids. However, during the campaign of Ottoman sultan Selim I over Safavids in 1514, he was careful not to provoke the Ottomans. In return Ottomans didn't annex any part of Afshar territory. Nevertheless, Sevindik, being Kızılbash renewed Safavid alliance after the death of Selim I. End of the beylik Ottomans decided to end the beylik. In 1534, during the 6th (Irakeyn) campaign of Suleyman I Pargalı İbrahim Pasha annexed the Afshar territory to Ottoman realm. In Erzurum a beylerbeylik (province) was established and a member of Dulkadir house was appointed as the beylerbey. References Category:Anatolian beyliks Category:States and territories established in the 1480s

Q: Print string over plotted line (mimic contour plot labels) The contour plot demo shows how you can plot the curves with the level value plotted over them, see below. Is there a way to do this same thing for a simple line plot like the one obtained with the code below? import matplotlib.pyplot as plt x = [1.81,1.715,1.78,1.613,1.629,1.714,1.62,1.738,1.495,1.669,1.57,1.877,1.385] y = [0.924,0.915,0.914,0.91,0.909,0.905,0.905,0.893,0.886,0.881,0.873,0.873,0.844] # This is the string that should show somewhere over the plotted line. line_string = 'name of line' # plotting plt.plot(x,y) plt.show() A: You could simply add some text (MPL Gallery) like import matplotlib.pyplot as plt import numpy as np x = [1.81,1.715,1.78,1.613,1.629,1.714,1.62,1.738,1.495,1.669,1.57,1.877,1.385] y = [0.924,0.915,0.914,0.91,0.909,0.905,0.905,0.893,0.886,0.881,0.873,0.873,0.844] # This is the string that should show somewhere over the plotted line. line_string = 'name of line' # plotting fig, ax = plt.subplots(1,1) l, = ax.plot(x,y) pos = [(x[-2]+x[-1])/2., (y[-2]+y[-1])/2.] # transform data points to screen space xscreen = ax.transData.transform(zip(x[-2::],y[-2::])) rot = np.rad2deg(np.arctan2(*np.abs(np.gradient(xscreen)[0][0][::-1]))) ltex = plt.text(pos[0], pos[1], line_string, size=9, rotation=rot, color = l.get_color(), ha="center", va="center",bbox = dict(ec='1',fc='1')) def updaterot(event): """Event to update the rotation of the labels""" xs = ax.transData.transform(zip(x[-2::],y[-2::])) rot = np.rad2deg(np.arctan2(*np.abs(np.gradient(xs)[0][0][::-1]))) ltex.set_rotation(rot) fig.canvas.mpl_connect('button_release_event', updaterot) plt.show() which gives This way you have maximum control. Note, the rotation is in degrees and in screen not data space. Update: As I recently needed automatic label rotations which update on zooming and panning, thus I updated my answer to account for these needs. Now the label rotation is updated on every mouse button release (the draw_event alone was not triggered when zooming). This approach uses matplotlib transformations to link the data and screen space as discussed in this tutorial. A: Based on Jakob's code, here is a function that rotates the text in data space, puts labels near a given x or y data coordinate, and works also with log plots. def label_line(line, label_text, near_i=None, near_x=None, near_y=None, rotation_offset=0, offset=(0,0)): """call l, = plt.loglog(x, y) label_line(l, "text", near_x=0.32) """ def put_label(i): """put label at given index""" i = min(i, len(x)-2) dx = sx[i+1] - sx[i] dy = sy[i+1] - sy[i] rotation = np.rad2deg(math.atan2(dy, dx)) + rotation_offset pos = [(x[i] + x[i+1])/2. + offset[0], (y[i] + y[i+1])/2 + offset[1]] plt.text(pos[0], pos[1], label_text, size=9, rotation=rotation, color = line.get_color(), ha="center", va="center", bbox = dict(ec='1',fc='1')) x = line.get_xdata() y = line.get_ydata() ax = line.get_axes() if ax.get_xscale() == 'log': sx = np.log10(x) # screen space else: sx = x if ax.get_yscale() == 'log': sy = np.log10(y) else: sy = y # find index if near_i is not None: i = near_i if i < 0: # sanitize negative i i = len(x) + i put_label(i) elif near_x is not None: for i in range(len(x)-2): if (x[i] < near_x and x[i+1] >= near_x) or (x[i+1] < near_x and x[i] >= near_x): put_label(i) elif near_y is not None: for i in range(len(y)-2): if (y[i] < near_y and y[i+1] >= near_y) or (y[i+1] < near_y and y[i] >= near_y): put_label(i) else: raise ValueError("Need one of near_i, near_x, near_y")

--- abstract: 'We present surface photometry and stellar kinematics of NGC 4264, a lenticular galaxy in the region of the Virgo Cluster undergoing a tidal interaction with its neighbour, NGC 4261. We measured the bar radius and strength from SDSS imaging and the bar pattern speed from MUSE integral-field spectroscopy. We find that NGC 4264 hosts a strong and large bar, which is rotating fast. The accurate measurement of the bar rotation rate allows us to exclude that the formation of the bar was triggered by the ongoing interaction.' author: - 'Virginia Cuomo$^{1,*}$, Enrico Maria Corsini$^{1,2}$, Alfonso J. L. Aguerri$^{3,4}$' - 'Victor P. Debattista$^5$' --- Introduction ============ The bar pattern speed [$\Omega_{\rm bar}$]{} is the angular velocity of the bar rotating around the galactic centre. It can be parametrised through the bar rotation rate [${\cal R}$]{}, which is defined as the ratio between the corotation radius [$R_{\rm cor}$]{} and the length of the bar semi-major axis, [$a_{\rm bar}$]{}. In turn, [$R_{\rm cor}$]{} is measured as the ratio between the circular velocity [$V_{\rm circ}$]{} of the galaxy and [$\Omega_{\rm bar}$]{}. Bars with $1.0 \leq {\cal R} \leq 1.4$ end close to their [$R_{\rm cor}$]{} and are called [*fast*]{}, while bars with ${\cal R} > 1.4$ are shorter than [$R_{\rm cor}$]{} and are called [*slow*]{} ([@Athanassoula1992; @Debattista2000 Athanassoula 1992, Debattista & Sellwood 2000]). The value of [$\Omega_{\rm bar}$]{} is expected to decrease with time because of the angular momentum exchange between the galaxy components and dynamical friction exerted on the bar by the dark matter (DM) halo. A huge amount of DM can efficiently slow down the rotation of the bar, which is expected to be [*slow*]{} ([@Debattista1998; @Athanassoula2013 Debattista & Sellwood 1998, Athanassoula et al. 2013]). The determination of [${\cal R}$]{} allows both to investigate the secular evolution of barred galaxies and to put constraints on the DM distribution in their central regions. The only direct way to recover [$\Omega_{\rm bar}$]{} is to apply the model-independent Tremaine-Weinberg method ([@Tremaine1984 TW, Tremaine & Weinberg 1984]). It is based on the simple equation $\Omega_{\rm bar}\sin i=\langle V\rangle/\langle X\rangle$, where [$\Omega_{\rm bar}$]{} is given by the ratio of the luminosity-averaged position $\langle X\rangle$ and line-of-sight (LOS) velocity $\langle V\rangle$ of the stars. These quantities are measured along apertures crossing the bar and located parallel to the disc major axis, when its inclination $i$ is known. The resulting values are fitted with a line and the corresponding slope is proportional to [$\Omega_{\rm bar}$]{}. Despite its simple formulation, the method has some restrictions: undisturbed morphology and kinematics are required to correctly identify the disc position angle (PA) along which to locate the apertures, so interacting galaxies are generally avoided. About 100 galaxies have been analysed with the TW method using long-slit ([@Corsini2011 Corsini 2011]) and integral-field spectroscopy ([@Debattista2004; @Aguerri2015; @Guo2019; @Cuomo2019 Debattista & Williams 2004, Aguerri et al. 2015, Guo et al. 2019, Cuomo et al. 2019a]), hosting only [*fast*]{} bar. No trends were found between [${\cal R}$]{} and Hubble type or DM content, but so far measurements have large uncertainties ($\langle\Delta\Omega_{\rm bar}/\Omega_{\rm bar}\rangle=0.3, \langle\Delta{\cal R}/{\cal R}\rangle=0.4$). NGC 4264: an interacting galaxy =============================== ![Photometric decomposition of NGC 4264. Upper panels: observed, modelled, and residual surface brightness distributions. The FOV is oriented with North up and East left. The apertures are shown as black lines in the observed image. Lower panels: radial profiles of surface brightness, PA, and ellipticity of the observed (black dots with gray error bars) and seeing-convolved modelled image (green solid line) and their difference. The surface brightness radial profiles of the best-fitting bulge (blue dashed line), bar (magenta dash-dotted line), and disc (orange dotted line) are shown for the semi-major axis distance to the galaxy centre. The red solid and dashed segments give the mean values of PA and ellipticity for the inner and outer portions of the disc, respectively.[]{data-label="fig:1"}](cuomo_fig1.eps) We analysed the case of NGC 4264, a lenticular galaxy in the region of the Virgo Cluster, with some hints of interaction. In fact, NGC 4264 is located close to the bright elliptical galaxy NGC 4261 at a projected distance of 30 kpc and they are probably gravitationally bound. We measured the surface photometry from the SDSS $i$-band images with the [iraf ellipse]{} task ([@Jedrzejewski1987 Jedrzejewski 1987]). We observed a twist in the outer disc isophotes. The PA radial profile (Fig. \[fig:1\], bottom-right panel) shows two disc regions characterised by two constant PAs, which differ by $\Delta{\rm PA}\sim10^{\circ}$. This is confirmed by the photometric decomposition we performed using the [gasp2d]{} code ([@Mendezabreu2014 M[é]{}ndez-Abreu et al. 2014]), in which we included a double-exponential law ([@Mendezabreu2017 M[é]{}ndez-Abreu et al. 2017]) to describe the contribution of the disc to the galaxy surface brightness (Fig. \[fig:1\]). Characterising the bar ====================== The correct determination of the disc PA is crucial for a solid measurement of $\Omega_{\rm bar}$ with the TW method ([@Debattista2003 Debattista 2003]). In order to deal with this problem in the case of an interacting galaxy, it is necessary to use the synergy between high-quality photometry and integral-field spectroscopy. The high spatial resolution of MUSE allowed us to define [*a posteriori*]{} the PA of the disc, and to fine-tune the number and length of the apertures used in the TW analysis. To characterise the bar it is necessary to measure its radius $a_{\rm bar}$, strength $S_{\rm bar}$, and pattern speed $\Omega_{\rm bar}$. We obtained $a_{\rm bar}=3.2\pm0.5$ kpc by applying three different methods to the SDSS $i$-band image: the photometric decomposition (Fig. \[fig:1\]), Fourier analysis, and analysis of the PA of the deprojected isophotes. The Fourier analysis ([@Aguerri2000 Aguerri et al. 2000]) allowed us also to derive $S_{\rm bar}=0.31\pm0.04$. Our results are consistent with typical values found for lenticular galaxies ([@Aguerri2009 Aguerri et al. 2009]). Finally, we measured $\Omega_{\rm bar}=71\pm4$ km s$^{-1}$ kpc$^{-1}$ with the TW method: we defined 9 apertures located parallel to the major axis of the inner disc. To recover the values of $\langle X\rangle$ and $\langle V\rangle$, we collapsed the MUSE datacube between 4800 and 5600 Å in the spectral and spatial directions, respectively. In particular, we extracted the LOS velocity from the resulting spectra. We fitted with the [idl fitexy]{} task the resulting values (Fig. \[fig:3\], red solid line). To recover $V_{\rm circ}=189\pm10$ km s$^{-1}$, we measured the stellar kinematics from the entire MUSE datacube with [ppxf]{} ([@Cappellari2004 Cappellari & Emsellem 2004]) and [gandalf]{} codes ([@Sarzi2006 Sarzi et al. 2006]) on a Voronoi tessellation ([@Cappellari2003 Cappellari, & Copin 2003]), and we applied the asymmetric drift correction, following the prescription of [@Aguerri2015]. Finally, we derived $R_{\rm cor}=2.8\pm0.2$ kpc, and ${\cal R}=0.88\pm0.23$. This means NGC 4264 hosts a [*fast*]{} bar. ![Pattern speed of the bar in NGC 4264 for the different tests performed to assess how the slope of the best-fitting straight line changes as a function of the aperture width (top left panel), PA (top right panel), wavelength range (bottom left panel), and number of apertures (bottom right panel). The reference value in each panel is given by the solid red line.[]{data-label="fig:3"}](cuomo_fig2.eps) Discussion and conclusions ========================== We tested the stability of our results by verifying all the ingredients of the TW analysis. We varied width, number, and spectral range used to define the apertures and we found compatible results. On the contrary, changing the PA of the apertures led to different results (Fig. \[fig:3\]). In particular, the PA of the outer disc corresponds to ${\cal R}<1.0$, which is unphysical. These tests confirm that the bar of NGC 4264 is living in the inner part of the disc, and that the correct identification of the PA of the disc is crucial for the TW method. All the details are given in [@Cuomo2019b]. This represents a pilot study for further accurate MUSE TW measurements: the values of [$\Omega_{\rm bar}$]{} and [${\cal R}$]{} for NGC 4264 are amongst the best-constrained ones ever obtained with the TW method ($\Delta\Omega_{\rm bar}/\Omega_{\rm bar} = 0.06$, $\Delta{\cal R}/{\cal R} = 0.26$). These results allowed us to constrain the formation mechanism of the bar in NGC 4264. We interpreted the twist of the outer isophotes as suggestive of a warp due to the ongoing interaction with NGC 4261. Since the bar of NGC 4264 is [*fast*]{}, we concluded that its formation was not triggered by the recent interaction or by a previous interaction with an other galaxy, because this would form a [*slow*]{} bar ([@MartinezValpueata2017 Martinez-Valpuesta et al. 2017]). Further accurate measurements of [$\Omega_{\rm bar}$]{} and [${\cal R}$]{} in a large number of barred galaxies would be of considerable interest to severely test the predictions of numerical simulations about the time evolution of [$a_{\rm bar}$]{} and [$\Omega_{\rm bar}$]{} as a function of gas content, luminous and DM distribution ([@Athanassoula2013 Athanassoula et al. 2013]). 2000, [A&A]{}, 361, 841 2009, [A&A]{}, 495, 491 2015, [A&A]{}, 576, A102 1992, [MNRAS]{}, 259, 345 2013, [MNRAS]{}, 429, 1949 2003, [MNRAS]{}, 342, 345 2004, PASP, 116, 138 2011, Memorie della Societa Astronomica Italiana Supplementi, 18, 23 2019, [A&A]{}, in press, (arXiv:1909.01023) 2019, [MNRAS]{}, 488, 4972 1998, [ApJ]{}, 493, L5 2000, [ApJ]{}, 543, 704 2004, [ApJ]{}, 605, 714 2006, [MNRAS]{}, 366, 1151 2003, [MNRAS]{}, 342, 1194 2017, [MNRAS]{}, 464, 1502 2014, [A&A]{}, 572, A25 2017, [A&A]{}, 598, A32 2019, [MNRAS]{}, 482, 1733 1987, [MNRAS]{}, 226, 747 1984, [ApJ]{}, 282, L5

Function and regulation of aflJ in the accumulation of aflatoxin early pathway intermediate in Aspergillus flavus. aflJ resides within the aflatoxin biosynthetic gene cluster adjacent to the pathway regulatory gene aflR and is involved in aflatoxin production, but its function is unknown. Over-expression of aflJ in the aflatoxin-producing strain 86-10 resulted in increased aflatoxin. In an effort to study the function and regulation of aflJ, strain 649-1 lacking the entire biosynthetic cluster was transformed with either reporter constructs, expression constructs, or cosmid clones and analysed for gene expression or metabolite accumulation. Over-expression of aflJ did not result in elevated transcription of ver-1, omtA or aflR. To determine if over-expression of aflJ leads to an increase in early pathway intermediates, strain 649-1 was transformed with cosmid 5E6 and either gpdA::aflJ alone, gpdA::aflR alone, or aflJ and aflR together. Cosmid 5E6 contains the genes pksA, nor-1, fas-1, and fas-2, which are required for the biosynthesis of the early pathway intermediate averantin. 649-1 transformants containing 5E6 alone produced no detectable averantin. In contrast, 5E6 transformants with gpdA::aflR produced averantin, but only half as much as those transformants containing both aflR and aflJ. Northern blot analysis showed that 5E6 transformants containing both aflR and aflJ had five times more pksA transcripts and four times more nor-1 transcripts than 5E6 transformants containing gpdA::aflR alone. Further, aflJ transcription was regulated by aflR. Over-expression of aflR resulted in elevated aflJ transcription. aflJ appears to modulate the regulation of early genes in aflatoxin biosynthesis.

Corey Haim has died at the tender age of just 38 after suffering what is being described as an “accidental overdose.” Haim was found unconscious by his mother at her apartment in California, and rushed to hospital in Burbank. Unfortunately, he was pronounced dead on arrival. This video is a fan-made tribute to the actor and recalls a happier time in his life when he was one of a new breed of child and teen actors on the scene. The cause of death hasn’t officially be announced yet but it’s believed to be due to an accidental overdose, with four bottles of prescription drugs found recovered at the scene. Haim is known to have had serious drink and drug addiction problems over the years, allegedly entering rehab 15 times. But it looks as though it’s a combination of legal medicine which got him in the end. Haim will hopefully be remembered for the good times rather than the bad, with memorable performances in films such as The Lost Boys being his lasting legacy.

Howard, Squires named Athletes of the WeekAll-Canadian skips leads curling teams to pair of CIS bronze medals in Regina WATERLOO, Ont. (March 25, 2014) - Women's curling skip Carly Howard of Penetanguishene, Ont., and men's curling skip Aaron Squires of St. Thomas, Ont., have been named the Laurier Bookstore Athlete of the Week for the week ending March 23, 2014. Howard took home CIS Second Team All-Canadian honours after helping the Golden Hawks capture a bronze medal at the 2014 CIS/CCA Curling Championships in Regina. The third year geography major was the second best skip statistically all week and guided Laurier to a 5-2 round robin record. Howard and the Hawks would wind up earning the bronze medal after losing to the eventual champions, the Carleton Ravens, in the semifinals. The medal is the sixth all-time for the Golden Hawks in the seven year history of the championship. Howard earns Female Athlete of the Week honours for the third time in her career, and third time this season. Squires was named a CIS First Team All-Canadian after leading the Golden Hawks to a bronze medal at the 2014 CIS/CCA Championships last week in Regina. The third year communication studies major took home first team honours at the skip position after posting the highest stats all week of competing skips. His play helped Laurier finish the round robin with a 6-1 record and the bronze medal after losing to Alberta in the semifinals. The third place finish is the highest for the Hawks since winning the inaugural championship seven years ago. This marks the first time in his career that Squires has been named the Male Athlete of the Week.

Q: Does Bruce Wayne have a middle name? Does Bruce Wayne have a middle name? I have heard many different ones so I want to know if he has a middle name and what it is. A: According to a web article from Bleeding Cool, published April 5, 2017, by Rich Johnston, titled "Lois Lane's Last Name And Bruce Wayne's Middle Name, Both Revealed Today (SPOILERS)", since the DC Rebirth (May 25, 2016–November 29, 2017), Bruce Wayne's full name now is apparently (now anyway), Bruce Thomas Wayne. Batman (Rebirth) #20 A: Bruce Wayne does not have a known middle name, but there are some common propositions for a middle name among them are Bruce, Robert, Anthony, Alen, or Thomas. Bruce: Possible if his real first name is Robert. Robert: Coming from Robert the Bruce, one of 2 names Bruce was named for by Bob Kane. Anthony: Mad Anthony Wayne, the other of the 2 names. Alen: His Grandfather's first name. Thomas: His father's first name. Thomas, Jr. was used in several places as a name for a Brother to Bruce. So Thomas is the least likely. Alen is likely due to him knowing a bit about Alen Wayne which means he was likely told as a child and usually that comes from parents saying "you were named after x. That's how you got your name"

/* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ package java.io; /** * Thrown when a program asks for a particular character converter that is * unavailable. */ public class UnsupportedEncodingException extends IOException { private static final long serialVersionUID = -4274276298326136670L; /** * Constructs a new {@code UnsupportedEncodingException} with its stack * trace filled in. */ public UnsupportedEncodingException() { } /** * Constructs a new {@code UnsupportedEncodingException} with its stack * trace and detail message filled in. * * @param detailMessage * the detail message for this exception. */ public UnsupportedEncodingException(String detailMessage) { super(detailMessage); } }

Shipyards Shipyards are mammoth orbital installations capable of manufacturing and launching advanced spacecraft in a matter of weeks, as opposed to the years a project would take in conventional drydocks on a planet's surface. They were constructed in a method similar to the average space station - a Hub block would be launched into Space via either chemical rockets from a planet or a spacecraft dropping one off, which enabled construction workers and drones, should they be available, to gradually add on to the block with different modules to create a fully functioning station. The Shipyard block was popularly considered one of these modules.

Ace Combat 6: Fires of Liberation Core Features Local Co-Op Not Supported Online Co-Op 4 Players Combo Co-Op (Local + Online) Not Supported LAN Play or System Link Not Supported Co-Op Extras Co-Op Specific Content The Co-Op Experience We know you're not happy unless you're going Mach II with your hair on fire. Climb into the cockpits of the world's most impressive fighters and get ready to prove to the world you're the best of the best! Photo-realistic planes and combat zones drop you straight into the heat of the action as you scramble to defend your home. Command your allies in the air, on the ground, and in the water to help you hold the lines and press forward into enemy territory. Battle on up to six fronts at once as battles unfold in real time. Then, fly online as you compete against pilots from all over the world in massive, head-to-head dogfights! Community Rating 3.45 out of 5 stars.31 total votes. This page contains co-op info for Ace Combat 6: Fires of Liberation on the Xbox 360. All information about Ace Combat 6: Fires of Liberation was correct at the time of posting. Information is subject to change. If you see any errors please email us.

Self Hypnosis For SuccessAttract Wealth And Prosperity To achieve prosperity and fulfilment by using self hypnosis for success is not only possible but is - if you approach it with diligence and consistency - virtually guaranteed to succeed. This is because all success, whether it concerns finances, career, relationships or whatever, is due to the mindset of the individual concerned. Someone with a negative mindset will encounter serious difficulty in succeeding with anything of any significance which they set out to do. On the other hand it's quite well known that many successful athletes use a form of self hypnosis for success in order to get their mindset attuned to the positive vision and the feeling of success in fulfilling their aims. They are aware of the power and the significance of the mind itself and of the integral part that it plays with regard to being able to achieve their goals. In other words, it doesn't really matter how physically fit you are if your mind isn't working for you as well as your body, because the body on its own will very possibly fail in one way or another. This obviously applies to all areas of life, health, financial, relationships, self esteem etc, and hypnosis/self hypnosis has been shown to be one of the most effective methods of replacing a negative and doubtful mindset with a positive and confident one. Let's assume that your current desire for success relates to your financial position. Everyone is aware that there are powerful forces at work in the world today which are causing financial hardship on a wide scale. What possible use could something like self hypnosis for success be if you'd just lost your job and were perhaps facing the loss of your home? The answer to that very good and relevant question is simply this. Problems, difficulties, stress, challenges, whatever you want to call it, arise periodically and inevitably throughout life. And if you stop to think about it, you understand that it's not really so much the problem itself that's important, but how you react to and deal with it. And in order to react to and deal with a problem, difficulty or challenge positively and constructively - and thus ensure a much higher possibility of overcoming it and even perhaps ultimately benefit from it - you need to have a basically positive mindset. And this highly empowering transformation from having a negative mindset to having a positive one can be achieved through self hypnosis. A person with a negative mindset can find him or her self being overwhelmed by a quite small problem, whereas someone with a positive mindset can face a much bigger problem and not only overcome it but come through it stronger, smarter and wiser than before. So if you're facing financial hardship or just lack of financial success, the problem lies not with the financial hardship or lack of success itself, but in how you choose to react to it. If you allow yourself to think that your financial position is hopeless, that there's no way out let alone forward, then you are choosing to react negatively to the situation. No matter how hopeless a situation may appear to be, it's ultimately you who decides how you're going to react to and deal with it. If nothing else, simply ask yourself exactly what's to be gained by allowing yourself to sink into worry, anger, despair or whatever. Self Hypnosis For Success Self Hypnosis is a means or technique of utilizing the vast and untapped resources of the subconscious mind. Unless you're at the very point of actual death, there is no such thing as a hopeless situation. Your subconscious mind has within it the solution to any problem or difficulty, however insurmountable or intractable it may seem to your "normal" consciousness and way of thinking. Even if you're broke, in debt and with seemingly no prospects you still have and always will have deep within you the ability to transform yourself into a wealthy and financially independent individual. To deny this is to deny yourself. It's therefore a question of getting to and using that hidden, untapped resource within yourself to enable you to dispel negative self belief and replace it with a positive mindset and to provide you with the creative ideas, inspiration and intuition through which you will find your way along the path to success. That hidden, untapped resource - the subconscious - can be accessed and harnessed through self hypnosis for success. When someone says "it's all in the mind" they are, in a manner of speaking at least, perfectly correct. Everything about you and your life springs from and is strongly influenced by your "mind". Your financial success or failure comes about ultimately because that's where your own mind has taken you. Obviously circumstances of birth and upbringing can make a difference, but ultimately it's your own mind, your own personality and inner being that drives you towards whatever situation it believes you should be in. This is why some people who have been born and raised in financially advantageous circumstances become poor and some people born and raised in poorer circumstances become wealthy. So if your quality of life is suffering because of financial lack, start using self hypnosis for success to begin turning things around and set yourself on the sure path to where you really want to be. Spend some daily time over the next few weeks deeply relaxing your body and mind and absorbing positive wealth related suggestions and mental imagery. As the subconscious gradually begins to accept and absorb this you'll soon find your whole outlook and attitude changing for the better and wondering how you could ever have doubted your ability to succeed financially. Details of the Millionaire Mindset hypnosis download which can help considerably to facilitate this process are available through the following link. "Hi, I must admit that the help I've had with my stress problem through using the relaxation hypnosis download has been amazing. I knew within a few days of starting with it that I'd at last found the solution. My life is so much better now. Thanks so much". Pam UK

Pneumococcal vaccination coverage among persons ≥65 years--United States, 2013. Invasive pneumococcal disease is a major cause of illness in the United States, and rates are higher among persons ≥65 years. Pneumococcal vaccination has been recommended to adults ≥65 years since 1997. Data from the 2005-2013 Behavioral Risk Factor Surveillance System were analyzed. Weighted estimates of pneumococcal vaccination coverage were calculated by state and race/ethnicity and tests for linear trend were performed. In 2013, the median state vaccination coverage among adults ≥65 years was 69.5%, and coverage ranged from 61.9% in New Jersey to 75.6% in Oregon. Coverage overall among non-Hispanic whites (71.1%) was higher than coverage for non-Hispanic blacks (57.7%), Hispanics (51.9%), and non-Hispanic persons of other race (65.4%). Coverage increased from 2005 to 2013 overall and by racial/ethnic subgroups. Although pneumococcal vaccination coverage has improved in the past several years, coverage remains below the Healthy People 2020 target of 90% and racial/ethnic disparities exist.

List of newspapers in South Africa This is a list of newspapers in South Africa. In 2017, there were 22 daily and 25 weekly major urban newspapers in South Africa, most published in English. According to a survey of the South African Audience Research Foundation, about 50% of the South African adult population are newspaper readers and 48% are magazine readers. Print media accounts for about 19.3% of the R34.4bn of advertising money spent in the country. Newspapers by circulation National publications Beeld (in 5 of 9 provinces) Business Day City Press Daily Sun KwelaXpress Mail & Guardian News Everyday Naweek Beeld The New Age Rapport Soccer Laduma Sondag (in 6 of 9 provinces) The Sowetan Sunday Independent Sunday Sun Sunday Times Sunday World The Teacher Townpress Vuk'uzenzele The Zimbabwean Publications by province Mpumalanga DAILY SUN Ermelo Insight (Mpumalanga)|Ermelo Insig Kasi Express Community Newspaper News Everyday Ridge Times Standerton Advertiser Higvelder Ziwaphi 013NEWS Bulletin Excelsior News Gauteng The Citizen The Joburg Times The Jozi Chronicle The Laudium Sun The Lenasia Sun Pretoria News Pretoria News Weekend The Saturday Star SowetanSpotlight News (Gauteng Provincial monthly newspaper) The Star Super Saturday Citizen tame TIMESThe TimesTownpress NewspaperKwaZulu-NatalThe Daily NewsIlangaIlanga langeSontoThe Independent on SaturdayIsolezweIsolezwe ngeSontoIsolezwe ngoMgqibeloThe MercuryPostThe Rising SunSunday TribuneTabloid MediaUmAfrikaWeekend WitnessThe WitnessThe Zululand ObserverIsambane NewsFree StateArts Ya RonaDumelangNewsExpressFree State NewsFree State SunFree State TimesIndigo magIssueThe Media NewsSaterdag VolksbladThe Sports Eye VolksbladVistaThe WeeklyNorthern CapeDiamond Fields Advertiser (DFA) Postmasburg Register FiND iT in Kimberley (FiND iT) Eastern CapeDaily Dispatch and the Saturday DispatchGo!&ExpressGrocott's MailI'solezwe lesiXhosa The Herald Weekend Post Western Cape Breederivier Gazette Die Burger Die Burger Saterdag Cape Argus Cape Times Daily Voice District Mail Drakenstein Gazette Eikestadnuus Helderberg Gazette Hermanus Times Paarl Post Die Son Son op Sondag Stellenbosch Gazette Swartland Gazette Theewaterskloof Gazette Vukani Weekend Argus (Saturday edition) Weekend Argus (Sunday edition) Weskusnuus Weslander Worcester Standard Student/university publications Activate - Rhodes University The Conduit - Durban University of Technology IRAWA Post - University of the Free State Journalismiziko - Durban University of Technology Die Matie - University of Stellenbosch NMMYou - Nelson Mandela Metropolitan University Nux - University of KwaZulu Natal The Oppidan Press - Rhodes University Perdeby - University of Pretoria Speculum - Central University of Technology, Free State UJ Observer - University of Johannesburg Varsity - University of Cape Town Wapad - North-West University Vuvuzela - University of the Witwatersrand Online only publications ANC Today Daily Maverick The South African The Daily VOX Eyewitness News FiND iT in Kimberley GroundUp_(news_agency) Indian Spice Mamba Online MatieMedia - University of Stellenbosch Mayihlome News News Everyday Property Wheel News The Western Cape #TheWC Regional/community publications These newspapers only serve small regions, towns, or communities within larger cities, or have a small circulation. Limpopo Mpumalanga North West Defunct These newspapers are no longer published. Rayton News/Nuus The Friend Hoofstad The Lantern New African New Nation Nova Oggendblad Oosterlig Rand Daily Mail South Sunday Express ThisDay Die Transvaler Die Vaderland Vrye Weekblad Weekend Mirror The Zingari See also Media of South Africa List of South African media Xhosa language newspapers South African Audience Research Foundation (SAARF) Online newspapers in South Africa South African news websites References Bibliography External links Contact directory: media contacts, South African Government communications News and media, South African Government information SAARF website AfricaBib South Africa Newspapers Category:Newspapers published in South Africa

A Tale About a Tiger A Tale About a Tiger Private investigator Lydia Chin infiltrates New York’s Chinatown to stop a dealer in valuable, and illegally obtained, animal parts–the sexual organ of the male tiger, to be exact–believed to have miraculous powers in the mysterious world of Chinese medicine. The tiger, in danger of becoming extinct, is revered by the Chinese people. And Lydia Chin will do her best to stop the criminals in question.

In view of the importance of sealing properties, elastic bodies such as rubber are used in parts which slide while maintaining their sealing performance, for example a gasket which is integrated with a syringe plunger and forms a seal between the plunger and barrel. Unfortunately, such elastic bodies have a slight problem with the sliding properties (see Patent Literature 1). Thus, a sliding property improving agent, for example silicone oil, is applied to the sliding surface. However, a concern has been raised over the potential adverse effects of silicone oil on recently marketed bio-preparations. On the other hand, gaskets not coated with a sliding property improving agent have poor sliding properties and thus do not allow plungers to be smoothly pushed but cause them to pulsate during administration, leading to problems such as an inaccurate injection amount and infliction of pain on patients. To satisfy the conflicting requirements, sealing properties and sliding properties, a coating technique using a self-lubricating PTFE film has been proposed (see Patent Literature 2). Unfortunately, such PTFE films are generally expensive and increase the production cost of processed products. Thus, the range of applications of these films is limited. Also, products coated with PTFE films might not be reliable when they are used in applications where sliding or the like is repeated and thus durability is required. Furthermore, since PTFE is vulnerable to radiation, the PTFE-coated products unfortunately cannot be sterilized by radiation. Consideration may also be given to the use in other applications where sliding properties are required in the presence of water. Specifically, water can be delivered without a loss by reducing the fluid resistance of the inner surface of a pre-filled syringe or of the inner surface of a pipe or tube for delivering water, or by increasing or markedly reducing the contact angle with water. Reduction in the surface resistance of the internal/external surfaces of a catheter tube facilitates insertion of the catheter into the body or introduction of a guide wire through the catheter. Drainage of water on wet roads and of snow on snowy roads can be improved by reducing the fluid resistance of the groove surfaces of tires, or by increasing or markedly reducing the contact angle with water. This results in enhanced grip performance and improved hydroplaning performance and therefore better safety. In addition, less adhesion of dirt and dusts can be expected when the sliding resistance of the sidewall surfaces of tires or the walls of buildings is reduced, or when their contact angle with water is increased. Further advantageous effects can be expected, including, for example: less pressure loss upon delivering water, an aqueous solution or the like through a diaphragm such as a diaphragm pump or valve; easy sliding of skis and snowboards achieved by enhancing the sliding properties of the sliding surfaces thereof; better noticeability of road signs and signboards achieved by enhancing the sliding properties thereof to allow snow to readily slide on the surface; reduction in water resistance or drag on the outer peripheries of ships and less adhesion of bacteria on the outer peripheries, achieved by reducing the sliding resistance of the outer peripheries or by increasing the contact angle with water; and reduction in water resistance or drag of swimsuits achieved by improving the sliding properties of the thread surfaces thereof.

Andre Agassi is at the US Open tonight, and despite retiring just three years ago, says the sport is being played at a different level, guys are faster and stronger, and "every ball is just hit so hard". He's also just amazed that the guy playing Roddick, listed at 5-9, 150 (Pat McEnroe just guessed he was 5-6 1/2), just hit a serve 126 mph, Andre said he is reminded all the time why he retired when he did. Compare that humbleness to all the baseball analysts, who nonsensically talk about "back in my day" and "the good old days", despite ample evidence that the player today throw harder, are in far better shape, and are far better than the players of a decade ago, never mind 30. I used words such as "soft" to describe JJ when he was sent down, and he comes off as even moreso now...working so much he was tired, hearing whispers, it just comes off as so whiny...I mean, this is a 27 year-old, right? This isn't a teenager who hasn't yet figured out the sun will rise and set regardless tomorrow. I hope he just kills the ball in September, a patented Hardy hot streak, and ups his trade value. We have seen the future, and it ain't named JJ. EDIT: I mean, I know 68 AB's is a small sample, so I'm not going to rip him for his Nashville numbers, which are pedestrian at best, but 3 walks? Seriously? It's not hard to look at that and know he has been swinging at terrible pitches, and is not going to have success in the majors if that continues. I'd look at each series and let him play versus the weakest RHP starting...Alcides hits lefties well, and needs confidence, but any way they can artificially inflate Hardy's numbers would considerably up his value. He also only had 8 K's, which tells me he just went up there swinging, and did not work the count much. Sadly, a cursory glance at Hardy's stats tells me he did not give a whole lot of effort in his 2.5 weeks in AAA. Michael Brantley entered the weekend hitting .267 (five HRs, 20 doubles, 35 RBI) at Columbus. The Indians probably were hoping for a higher batting average and OPS (.703), but they are not concerned about the 22-year-old. They see steady progress. Here are his batting averages by month: .217, .261, .270, .280 and .304. He had an ankle injury that bothered him for several weeks, yet still has stolen 42 bases in 47 attempts. "His defense in center field has been off the charts," said Atkins. "He is going to be a very good player."--Cleveland.com First of all, I heard from every would be scout (you know, Braun has a hitch in his swing, Hart's elbow sticks out too far, Fielder will have to be a DH) in existence that Brantley could never play CF, and when I said "He has good speed and a decent arm, I don't know why he couldn't", the response was, "He can't". Why do I bother? Am I the only person who "gets" the fact that the minor leagues are for players to learn how to do certain things, and many may not be good at it right away? They hint at it, but a 22 year-old in AAA is very young, so those numbers are not bad at all (they aren't as good as Escobar's though, FYI, who is also 22). He may not be ready to open '10, but I bet he'll be ready by mid '10. With his speed and athletic ability, he may well end up being more of a loss than Laporta, who will all but for sure contribute 25 homers and 90 RBI's as soon as next year. And yes, I know there are no playoffs without CC. That said, it's so hard giving up 6 years of those kids. In Brantley's case, I'd have no problem going with a Gerut/Jason B platoon if Mike was waiting at AAA, likely to be ready soon. As is, Lo Cain, after a wasted '09, looks like he's a minimum of a year away, and quite likely more...you can't just look for a short-term solution. I'll be in the minority here, but I'd be all over Linebrink, especially if the Sox just want to dump salary and are only asking for a midlevel prospect. Scott is not cheap, but he'd provide solid depth as well as a backup closer in case Hoffman retires or decides to go elsewhere. Especially with Riske in doubt for next year, I'd love to add him to the '10 picture. He'll make $5M next year and $5.5M in '11, so I would not give them much at all for him, as he is a bit overpaid, but he has had about as solid of a career as any reliever you'll find. EDIT: I just went over to BR to get the link for his career stats, and I see he has had a terrible 2nd half...over an 8 ERA. This is a great time to get a great deal on him, but when you're a small market team with a limited budget, it's a big risk to take. Color me a lot less gung ho after seeing his recent stats. Numbers like that are almost always indicative of an injury of some type. Jon Heyman is reporting interest in Mike Cameron as the deadline nears. I'm surprised he and Counsell are not both headed elsewhere, I'm not sure what a contender could want if not solid defenders who are also respected vets, and average bats as well. If the Crew is not willing to offer Cameron arby (and they may well not, as he is old for a CF, and making $10M this year), they should trade him for anything, as something is always better than nothing. I'd love to see Corey Hart play some CF between now and game 162, to see if that's a possibility for the future. I know he's played there a bit, but I can't say I recall anything except him catching a few routine fly balls. He certainly has the speed and arm, but needless to say, he'd be a notch or three below Cameron. I'm a pretty causal tennis fan, having enjoyed it more seriously back when Agassi and Sampras and other Americans were on top of the circuit. I still get Tennis magazine (though it's free for me), however, and I read a while ago that the US Open is on ESPN2 now, after decades on USA. From what I see, it will be on every single day (except maybe the weekends) from noon to 10/11PM Central time, and even later, should a late night match go long. Let me tell you, there is no better tennis atmosphere in all the land than a 5 setter about 1AM New York time. There's no "taking plays off" or switching to a zone", it's one-on-one, gut wrenching action. I dare you to watch it for a while and then not stay up to see how it ends. Friend of Ramblings Jeff Sackmann discusses the qualifying for the Open (which I've always heard is played even harder than the Open, because if you don't qualify, you don't make enough to pay for your expenses, as well as who may win at his Summer of Jeff weblog. Philip sends this JS piece on literally millions of dollars being thrown down the drain by a single fraudulent day care and the inept watchdogs (or lack thereof) put in place by a clearly inept governmental agency. They should blow it up and just start over. This single woman has probably scammed a minimum of $5M from the state and the taxpayers, and there are what...thousands of such daycares? How pathetic. If the gov't in charge of it, it will, without a doubt, be poorly run. Kudos to the JS. In a little while, these folks may well be deciding whether or not you can have that medical procedure...or not. Good luck with that. NFLShop.com, the primary sales vehicle for National Football League jerseys, reports that Brett Favre of the Minnesota Vikings had the top-selling jersey in the time period from April 1 through Aug. 28. Jay Cutler, who also switched teams, from the Denver Broncos to the Chicago Bears, was second in the jersey rankings. Aaron Rodgers of the Green Bay Packers, who didn't switch teams, broke into the top 20, ranking 19th. No other Packer player had a jersey ranked in the top 25. Michael Crabtree, a wide receiver who hasn't even signed with the San Francisco 49ers, ranked 20th. Interestingly, Michael Vick of the Philadelphia Eagles, who switched from prison to football, ranked 4th in jerseys sold.--Don Walker, JS --OK, first of all, Favre's jersey sold more in 2 weeks than any other jersey did in 5 months. Oh, to be selling Vikings' merch of late. As a former retail manager, I chuckle at next year, second half of August, when your sales are down 70%, and this conversation takes place: Corporate figurehead: Al, your sales are down 72% this week. Al: I know, all of MN is way down, remember, last year is when Favre signed... CF: Al, you know we don't make excuses in this company. Al: Yeah, but let's be realistic, last year, sales were up because traffic was up, not because average purchase was up. This year, traffic's way down, and we're suffering. In fact, the average purchase is up 8%, meaning we're doing a better job merchandising and selling. CF: Well, if you think a negative 72% is acceptable...what's that noise? Al: Just banging my head against the wall. --Secondly, the idea that Vick's is 4th makes me believe many of those were sold because of his criminal past, which makes me shake my head for the people buying it. --So many of the top ones seem to be either rookies or players that changed teams, which makes sense, I suppose, because if you like Bears' stuff, you may well have a Urlacher jersey, but not a Cutler. Gotta wonder if Lopez might get some bench time, as he is almost playing on one foot of late, especially with more bodies soon to be up. It'll be interesting if Hardy and Escobar ever end up on the field at the same time, and if they do, who plays SS and who plays elsewhere? I tell you what, a manager goes from an idiot to a genius mighty quick when you get 5-7 good innings from your SP, and can bypass your weakest members of the bullpen corps. If the rosters weren't expanding Tuesday, you'd likely see them go back to 12 arms and 13 position players. As is, I would assume a pitcher or two and a position player or two will be added before the next game, and then more as the AAA season concludes (and Hart's rehab ends, and Burns/Iribarren 10 day demotion period ends. Chris Smith has been outstanding in Nashville, so he is very deserving. Other favorites to return? I'd say JJ and Dillard. Since they made room on the 40 man for a spot, almost anyone could be up, who has played well. Gotta wonder if McGehee lost that ground ball in the sun/shade mix or maybe in the crowd...he just flat out missed it by 6-9 inches off to the side. Suppan is a nice vet to throw against an aggressive Pirates' club...they attack, and it plays right into Jeff's hands. They've got a lot of hits, but most of them have been just grounders through the 5/6 hole, and usually erased by double plays. The Brewers do not intend to recall shortstop J.J. Hardy from Triple-A Nashville before Sept. 1, and as a result, Hardy will not have the Major League service time he needs to qualify for free agency following next season. Instead, he'll have an extra arbitration year before reaching the open market in the winter of 2011---Adam, MLB.com Obviously, this gives JJ an extra year of team control, and makes him worth a lot more on the trade market. Manager Ken Macha met with Melvin on Friday afternoon to discuss September callups. The plan calls for three pitchers, two infielders and one outfielder, but those promotions will be staggered through the first week of September, because the Triple-A Nashville Sounds are bidding to make the playoffs. The Sounds' regular season ends Sept. 7. I would assume they will bring up a couple arms right away to eat up innings in lopsided games. If Angel Salome is injured, it'll be too bad if they do not add another catcher, so they could PH or PR twice with the C's. Also, I'd love to see Salome catch just to see where he's at with his defense. Hardy is obviously one of the infielders, and I assume Iribarren is the other. Chris Smith will be back, as will Mike Burns. Tim Dillard is a possibility, the only other I see on the 40 man. As for the OF, Corey Patterson was recently added and has done well. Brendan Katin is a slugger who doesn't walk, but plays good defense, so he's also a possibility. Another topic for Melvin and Macha was shortstop Alcides Escobar. The Brewers must decide this winter whether Escobar is ready to take the everyday job from Hardy. "I don't think the window has been big enough to judge Escobar yet," Macha said. Time will tell if this is true or not. Alcides is still very young, and may struggle at times with the bat (though he's been far better than I thought), but I've had him penciled in for '10 SS for over a year now, and still do. As long as they ensure they have a solid veteran backup (Counsell, or a similar utility type), SS is in good hands. UPDATE: I forgot about Corey Hart returning. Doubtful they add Patterson or Katin. I love Conor Jackson, and while I suppose ARI is worried he will not be 100% as he recovers from "valley fever", I'd take that risk and stick him out in LF, while handing Braun RF. Look at how consistent his numbers were at the link. ...Brewers manager Ken Macha said G.M. Doug Melvin "scoured the area for pitching" before the deadline but was unable to add a starter because he refused to trade Mat Gamel or Alcides Escobar. We've heard this before, but it sure seems to be the truth, based on pure repetition. All those teams love 'em for a reason, you know. Cardinals pitching coach Dave Duncan may leave the organization after the year. Duncan is upset at the perceived handling of his son, Chris Duncan, by both the Cardinals media and front office. Tony LaRussa's right-hand man since 1983 has also lost influence with the front office, no longer being consulted on decisions and not seeing his philosophy mesh with the minor league philosophy. Duncan insists any decision will come in the off-season and be a professional, not personal decision. Color us skeptical. Duncan is possibly the best pitching coach in the game now that Leo Mazzone has retired. He won't have trouble finding a job. It would be a blow to the Cardinals' future as fantasy owners would no longer be able to pick up Cardinal pitchers and see them morph into world-beaters on a regular basis.--Rotoworld Duncan has certainly had his share of success. Not sure why he'd go elsewhere over something so minor, but needless to say, on the open market, he might get more than many managers. If only Mike Moore would have lived 100 years ago, he could have hung around with Lenin and Stalin, and we could all laugh at him now...of course, we laugh at him now anyway...there's no quicker way to lose credibility than to glorify Cuba's healthcare system and then to argue the USSR had it right. Here's something I do not recall doing before...Brewers trivia. All answers are easily searchable, so the challenge here is to get them without using Goggle (or Baseball Reference). To quote Letterman; please, please, no wagering. 1. Who has played the most 2B this year? 2. Who has played the most 3B? 3. Lowest OBP of the following? A. Corey HartB. Jason KendallC. Mat Gamel 4. With a runner at 3B and less than 2 outs, which two regulars have got the runner in most often? 5. Prince has been walked intentionally the most...who is 2nd and 3rd? 6. How many times has the SP thrown 120 or more pitches this year? 7. Only one pitcher has allowed more ground balls than fly balls...who is it? 8. If the Brewers allow 6 runs or less, what win pace are they on? A. 100B. 110C. 120 9. If the Brewers score 4 runs or more, what win pace are they on? A. 88B. 96C. 105 10. The Brewers have both allowed and scored the most runs in the same inning this year...which inning is that? The Cubs are more open to dealing Rich Harden to the Twins than you might think. The Cubs could receive two high draft picks if they offer Harden arbitration after the season and he signs elsewhere, but that's a risky strategy. Given his injury history, the club might not want to take a $10MM or so hit if he accepts, even for only one year.--Rosenthal, via MLBTR I can't imagine the Twins offering him arby, so they'll literally be trading for a half-dozen starts. Ryan Doumit is absent from Saturday's lineup after being benched in the fifth inning of Friday's game. Manager John Russell, visibly upset after the game, declined to elaborate Friday night. Perhaps we'll get some clarification at some point soon. Regardless of what transpired, this is likely only a short-term move to send a message; Doumit is still the starting catcher.--Rotoworld Gee, if the Pirates want to get rid of Doumit...allow me to dream for a minute. I missed this when it happened last night, I wondered why Jaramillo was batting in the 9th, as he is known for his defense. That said, though he's been bothered by injury, Doumit has been nothing compared to last year. John Stossel, the reason I attended, was enjoyable, to the point, and should have spoken longer. His 15 minutes was by far the highlight of the night, getting off several zingers (saying that ignoring pre-existing conditions was like Lindsay Lohan and himself paying the same for auto insurance), and telling well known stories of insuring his ocean front home with the only people dumb enough to do so (the gov't), and comparing how Visa does business to the gov't....brilliant stuff. The rest was just a bunch of capitalists showing how disgusted they are. I was surprised how old the audience was...most were retired, 65+, and I was among the younger 10% (I'm 40), though a lot of my age group and younger came late, and stood along side the chairs that were set up. As I told the Rambling wife, the Dems are lucky the Congressional elections aren't this November. I doubt the anger will last until November of '10, and tons of things will change before the election in 15 months. "Also, someone (either Dave Nelson or Chris Bosio, I think) mentioned tonight that Parra is nearing more innings than he's ever thrown...he's at 123 now, and he threw 166 last year...he's nowhere near where he's been just recently.--Ramblings" Is it possible that the team is counting the innings Parra threw inNashville when they are assessing his workload? Chris Ah yes, I forgot about his AAA time, and it makes perfect sense now. Thanks for the reminder, Chris. I doubt Manny will be limited too much the rest of the way, as it is usually considered "safe" if they add 25-30 innings a year, which should be well within Parra's normal pitching. I read that and thought to myself, "I'm sure it is". My gosh, our administration won't evn use the word "terrorist", in hopes they magically go away...CNN has to be a little left of that...and that's really far left. Neyer says it's a good deal for both sides, but he points out Rays fans probably won't like it. Dumping your face of the franchise, even in a down year, while 4.5 games out of the playoff hunt, is not going to be well received. Melvin did talk to two other teams about other players, but did not seem moved to trade any of his veterans before the Aug. 31 deadline to acquire players eligible for postseason rosters. "I don't think so," said Melvin, asked if he anticipated any trades. "But you don't know."--MLB.com I'm thinking a minor move or two will be made before August 31st. Guys like Counsell, Cameron, and so on are the epitome of what a contender looks for...good clubhouse guys, solid defenders, good OBP's. Reading between the lines (Adam said Doug has not been "available" the last 24 hours, and he hasn't even spoken with SF about Hoffman), it would seem he's been talking to somebody about someone. Also, someone (either Dave Nelson or Chris Bosio, I think) mentioned tonight that Parra is nearing more innings than he's ever thrown...he's at 123 now, and he threw 166 last year...he's nowhere near where he's been just recently. Doug says the same thing I have been...some of the improvements in pitching for next year will have to come from within. Certainly, Parra should improve some, and if Bush and Suppan are healthy, they should as well. Jason B not only gets his first HR, but is lucky enough to hit it into the bullpen, where selfish, greedy embarrassments like the Happy Youngster can't hold it for ransom. The TV team alludes to this whenever there is a 1st HR, but don't ever mention the specifics. If you are new here, check out Deadspin's recap of the shenanigans, and make sure you read the commenters rip him apart. Everyone I've known (mostly worked with) that has been a pothead is the exact same way as the article Glenn links to. The irony is, whenever I discuss this, I get e-mail that says, "You know Al, some college professors and artists and lawyers use pot, and they aren't bothered a bit". I always reply the same way...I'm sure some lawyers use meth too, and I wouldn't recommend that either. I still remember a gal I worked with in retail whose boyfriend was due to be sentenced for selling drugs, who was obviously a user. However, she was a manager, and she was closing the night after I opened...and despite being scheduled at 1, was not there at 4. She finally called at 4:30, and made it by 5, when I was supposed to leave. She made some excuse I would describe as "unbelievable", and I told her I'd see her later. She quit showing up the next week, so I've always been thankful she made it in that day and kept me from being there 14+ hours. We'll have to see how much box office the latest Woodstock movie does this weekend. If there is anything that's been overly discussed, other than the Kennedys, it's Woodstock. I can't imagine teens and young people today spending $9 to see a movie about this overhyped event. Its former numbers are difficult to estimate but one ornithologist, Alexander Wilson, estimated that one flock he observed contained over two billion birds.--Wiki Wilson was speaking of a flock of passenger pigeons, which almost beyond belief, were extinct 35 years after he saw that huge group. While 2B seems fictional, I have read tales of flocks so big they blocked the sun, and literally turned day into evening, back in the 1800's. I found this while looking up what a "mourning dove" looked like, as hunting season began for them this week. Turns out they are the most sought after game bird in the world...who knew? I can't believe so many adults wear bike helmets, as they look very uncomfortable and kinda goofy. Given I rode in the car on my mom's lap as a baby, never wore a helmet, and still rarely use a seat belt, it is astounding I'm still a member of society. It also amazes me so many motorcycle riders do not wear a helmet, but I do believe it is their choice. Being a believer in gov't shutting up and letting people make their own choices, as long as they do not infringe upon other's rights, makes my stances consistent, at least. It's been a disappointing season for (Milton) Bradley, but he has still managed a .387 OBP.---MLBTR I commented on this earlier, but Bradley's eye is still superb, and his reduction in power numbers is all but certainly due to an injury. That said, he has always been fragile, and may be unable to stay healthy while playing OF regularly. Doug has always been a guy to resist anything but "good clubhouse guys" as well, though I know several have said Milton is fine except in the heat of the moment. I tell you what...if he had a single year left, I'd see what the Cubs wanted. However, with 2 guaranteed years at $21M total, I wouldn't...unless they wanted to swap salary somehow. I can't say I've ever seen his show (other than flipping past it), but being I'm for, you know, freedom of speech, I can't say I'm worried about what he says. Some folks seem unaware remotes have a channel change button. Just read your post about Playboy's summation of the 2012 hysteria, and it brought to mind a terrific book by Charles Panati called the "Browser's Book of Endings: The End of Everything" that examines history and details (among other things) how different species, people, religions, and civilizations have ended. There's a chapter called "The End of the World" that's available for free online that you might find interesting. Suffice to say, the idiots who think the world will end in 2012 are certainly not the first and will certainly not be the last. Let me make sure I get this straight...ABC ran a day long, non-compensated infomercial thaqt fawned over the Obamacare...but they will not run an ad critical of it...even though they'd be paid to do so. Luckily for us, our media is unbiased, so the public can make up their own mind. Jason sends along this link, telling how to get a cell phone free, "supported by the US government". Jason adds he did not think a cell phone was at all necessary, and he's correct, I still would probably not have one if not for the fact it makes my business easier to be able to return calls while between appointments. Obviously, if the gov't is giving this company a dime of taxpayer money, it's headshakingly stupid. However, in reading the fine print, they give you time for free, but you're able to buy more, so this might be a "win-win". It might actually just be a phone company challenging ethics...anyone might be able to get a free phone. The irony here is, casual fans ignore small samples, and all the "pundits" look at nothing but tiny samples all the time. Golfer has a bad 18th hole, he can't handle the pressure...never mind he had to play brilliantly to be in the battle for the lead. I'm torn on the cash for clunkers thing, while I'm against most all forms of gov't intervention such as that, but I also have little doubt the program did some "good" for the economy, though very short-term and probably accomplished little other than clearing the lots before the '10 models hit. John Stossel points out the pure lunacy of destroying quality vehicles...many of which would be great for people who could not afford a new car even with the $3500-4500. Heck, many of the vehicles being destroyed are better than my current drive...hmmm. When in doubt, lowering the middle class family's taxes and not spending money like idiots would accomplish far more than these "magic beans" solutions that keep getting thrown at us. If he is claimed, Lopez may have to prove himself healthy before a deal can be made. He sure has been beat up since his arrival, but to his credit, has played well and insisted upon being in the lineup. Brewers general manager Doug Melvin, however, is disinclined to make trades during the August waiver period, believing that he cannot receive adequate returns. Hoffman, in particular, would appeal to teams in need of late-inning relief. He lacks no-trade protection, but is comfortable in Milwaukee, is open to returning next season and opposed to a setup role, according to a source with knowledge of his thinking. Kendall probably stands a better chance of being moved; several teams are looking for experienced catching. Counsell, too, is a strong veteran presence who could prove an asset to a contending team.--Rosenthal This is what real writers do while others are posting fictional accounts of games from the press box a few feet away from home plate. I believe you are able to put either 6 or 7 players on waivers each day, and it seems to me that those players were chosen for a specific reason...all are free agents, or have '10 options, and the Brewers have decisions to make on each one of them. I realize what Doug is saying is correct, but if they know they are not going to offer a certain player (Cameron and Kendall are the two that jump out to me) arby, they will be in effect losing them for nothing. Would a platoon of Gerut and Jason B cost the team several wins in September? Unlikely, but they'd likely be a step down on both offense and defense. You'd have to balance out adding a piece or three for the future with that. 9th inning: Phillips and Rolen singled against Burns, but Nix popped up for the first out. Darnell McDonald grounded out to McGehee. Burns walked Miller to load the bases for Adam Rosales, who flew out to leave the bases loaded.--Tony W's game blog One does have to wonder how a human being of any intelligence at all can mistake Narveson, a LH, for Burns, a RH. I honestly do not know if he even watches the games...is he maybe just banging out the play-by-play as it's posted online? Cameron lined out to left, but Braun for some reason tagged on a ball that wasn't that deep and was easily thrown out at the plate to send the game into extras. Seriously, this guy played the game? It took a perfect throw to get Braun...and then, the tag was barely in time. The scripts for these episodes are chock-full of emotional resonance. But it's a very fine line between angsty introspection and indulgent moaning. Particularly when your characters are good-looking Ivy League graduates with perfect health and rewarding careers.--Slate, on the very easy to hate 30something, which is now on DVD I was just out of high school when this show was on, and found it dull, whiny, and incredible "look at me!". Over the years, I've met a lot of people who seemed to think their life was deserving of the angst and over thinking these unwatchable characters did...and I don't hang around them either. If I were to ever meet those fictional characters, I'd say to them the same thing I say to the idiots who still try to get the wave going at sporting events...sit down and shut the hell up. Remarkable comeback, though no one will care if the Reds win. Frighteningly bad luck, as Prince, Casey, and Cameron all hit it right to the LF, making his throws very easy. They couldn't do that in BP if they tried to. The retired general is right, but let's be honest...why should LA spend a penny on preparing for a hurricane when they can just sit back, spend their money on wine, women, and song, leave their school buses sitting idle, and beg the federal government for cash. You know, some people are still living in FEMA trailers ye these many years after Katrina. The idea of prepping is, dare I say, not really necessary in these "what can my country do for me" times. The Brewers are back in Milwaukee to start a six-game home stand against Cincinnati and Pittsburgh. The team is clinging to playoff hopes that may or may not be an illusion, and because of that Craig Counsell is playing shortstop against Reds right-hander Bronson Arroyo. This is Counsell's 104th game of the season, a number I'm sure fans hoped he wouldn't hit based on the success of the rest of the infield. But for one reason or another, he's there and playing tonight instead of Alcides Escobar. It seems like Escobar won't play everyday until the Brewers are mathematically eliminated.--Tony W, JS Tony has a way of writing that irritates me to no end, and this is one of them. He throws out opinions and pretends they're facts. I have no problem with Escobar being set up to succeed, as Arroyo is tough versus RHP. And, I have to believe Hardy will be utilized when he returns, so I doubt Alcides plays "every day" late in the season either. This is one of those things you run into, and while I certainly would not be picking up guys in trade that will only be around until late September, I'm not saying they should quit trying either. It means a lot to the players and the organization to win as many games as possible, and you run into ethical issues if you run a lineup out there that, well, sucks. The odd thing is, Escobar already has a ton of PA's, and will hardly be helped or hurt by playing a couple extra games. Simply put, young players should be set up to succeed, and that's exactly what they are doing with him. Sadly, Tony W, and many others, are either too dimwitted or too negative to "get it". Because of his recent troubles, the Rambling dog is on a "bland diet", which is boiled rice and ground beef. Of course, I would find this delicious with some seasoning and country gravy poured over the top, but though it's refrigerated and looks like people food, it probably lacks flavor. The funny thing is, as I told the Rambling wife while preparing the feast last night, 2 cups of rice and a half-pound of ground beef is really inexpensive, about $1.50, and if memory serves, will last 7-10 days. Sadly, I'm pretty sure it's cheaper than his regular kibble. Our vet technician, who is also our neighbor, told us we could add yogurt as well, which supposedly would help his stomach work properly again, but I have little doubt if I added moisture like that, I'd feel obliged to taste it. Rhode Island shuts down gov't offices for 12 days...I'd say that might just be a good start.:) The irony here is, if you can get by without these "vital" positions a dozen days, why not two dozen? Or, all the time? My old roommate once worked for a suburb and pointed out that when the gal that answered the phone was out sick or on vacation (she made sure she took every sick day, if you know what I mean), someone else had to cover for her, or several people would cover her for 1-2 hours a day. I've always wondered if she had any idea that she was just proving herself unnecessary. One day, the Rambling wife, subbing for a local school district, sat in an office area all day reading a book, unless a teacher came in and asked her to make copies or collate said copies. When I asked why this position had not been cut, as you always hear how tight the districts are now run, she just shook her head and wondered the same thing. Today in his chat, Anthony Witrado said he played minor league ball, and yet, I can find no record of his ever doing so? Your thoughts? Frankie I assume Tony W never played in a game, but went to spring training with SEA and FLA, the organizations he says he suited up for. Surely he's not dumb enough (insert your own punchline here) to think minor league stats can't be looked up. I just checked BR.com to see if Witrado might be a "pen name", but could find no player named Anthony who played for those organizations, though I could have missed one, Anthony being a rather popular name and all. By the way, Frankie, I admire your pain tolerance, just while looking up what was said, I came across several times that Tony W said something either that was not true or hearsay thrown out as factual. I should go down the chat and tear it apart, but I don't think I have the time. I'd say he was unreadable, but I wouldn't want to offend those that are unreadable. Iribarren will come up as soon as he's eligible, I would assume. Who the Crew brings up on 9/1 will depend on where they are in the race and where Nashville is, but I would expect 3-5 guys, a couple pitchers to eat innings in lopsided affairs and some position players. Add one more to the list of positions they need to fill when they get done saving the world from ourselves. The one thing I do not understand is the "doctors and ER's will be filled to capacity"...the numbers presented don't look that overwhelming at all. Again, would I put it past the administration to "create a problem" and then solve it? Not at all, but they look like they need to market the horror a tad better. UPDATE: I'm struggling with the math, especially this quote: As many as 300,000 patients may be treated in hospital intensive care units, filling 50 percent to 100 percent of the available beds There are about 6000 hospitals in the US, per Google. That would equal 50 per hospital...that's a lot. However, I don't think anyone is claiming all 50 per hospital would be all at once. If they had to stay a week each, that's 350 hospital days per, but spread out over 2 months...that would be about 6 patients a day. Given that the flu may last the entire winter, 350/120 is less than 3 per day. Not much movement in the role player realm this year, as most teams have either went big or done nothing. In theory, I understand this, but when a playoff spot can easily be decided by a game or two, a slight improvement in a couple spots might be enough to do it. Any insight on what the Brewers are likely to do next year at catcher. You have said sometime earlier that it could be Salome splitting time with a veteran. Do you still feel that way? If so who might the veteran be? Perhaps Rivera? I assume they will part with Kendall...please tell me they will. And finally the rotation would be with today's roster some form of Gallardo, Parra, Bush, Looper and Suppan....and an upgrade is obviously needed who is the odd man out? I would speculate Bush as Looper and Suppan seem untradeable. Please help if you can. Clint First off, Looper has a mutual option, so the Crew can easily be rid of him. However, given how bad the rotation has been, getting rid of any SP who is considered a major leaguer can't be considered a plus. If nothing else, there are plenty of teams that would happily put Braden in their rotation for $6M, so he would have trade value. I'm not a big Rivera fan, and I would have said Kendall is likely to be back until a few weeks ago, but now, I'm not sure. His OBP has been decent, but his complete lack of even doubles power makes him a liability. He'll have to take a huge pay cut to return, without a doubt. UPDATE: I forgot to mention that if Kendall does return, that's a mighty weak 7/8/9 part of the lineup, with almost no power at all. Because Alcides is all but guaranteed to be in the '10 lineup, they may not want to have a pair of low SLG guys in the lineup...that makes it tough to keep Jason around. Looks like a lot of DH's will be available next season for those AL teams in need of one. Matsui would be passable in LF, and I guess Sheffield would too (though he has not aged gracefully), but for the most part, it is a list of DH's. The US currently has $7.4T in debt...that's the accumulation of all past deficits. Under Obama's proposed plan (which tend to be optimistic projections), his administration will add $9T in the next decade. Not only is today the last day of my vacation, the Rambling dog is having internal problems that can only be described as "smelly" and "disturbing". The Rambling wife just called to see if I contacted the vet, but I told her we might as well wait to see if it passes (no pun intended), though I think we'd both be willing to pay good money just to have them be in possession of Ronnie today. The last time he had this issue was last summer, when we found out he had lyme disease. It isn't identical symptoms, so I doubt that's it, not to mention he was vaccinated for it both years, which would seemingly make the odds of him getting it again very tiny. Whenever someone suggests cutting costs by letting criminals out of prison, I always say to myself, you know, they ain't in there for jaywalking...these are evil, bad men. Note this was only a medium security facility...the really bad guys are in maximum security. I would imagine there will suddenly be a very anti-prisoner movement in Kentucky, as there isn't any doubt that it will be rebuilt...and fireproof I assume. Even though the Brewers rotation (other than Yo) has struggled, the offense has kept them right in the race...except for when Parra and CV have drawn the start: Overall: 60-63MP & CV: 9-18Others: 51-45 Without Manny struggling through a crappy season, and CV being pressed into a role he was not prepared for, the Brewers are on a pace for 86 wins, just 4 less than last year. We'll see how the final quarter of the campaign goes, but despite the fact their ERA is high, the rotation has been serviceable (at least with this offense), minus the 27 times CV and Parra have toed the rubber. While many will call for a complete overhaul (which is not going to happen anyway, due to contractual obligations), by the looks of things, if Parra would have had a "normal" Manny year, or with the addition of a single league average pitcher, this '09 version could be right in the thick of the wildcard race, even with Burns and McClung having to pick up a few starts. Funny, I haven't been getting e-mails of late from people telling me how popular the president is...maybe this is why. The astounding part is, he started at about a +30, and is now at -14 in just a few months. Of course, this is with the old media minimizing his off-the-teleprompter gaffes...Jon Stewart would be running that Special Olympics statement every night if Bush had said it. There's plenty of time before '12, but for those looking at the '10 Congressional votes, it would have been hard to imagine this scenario back when it was cold out. Of course Braun was safe at 1B...on a close call, the umpire is pretty much always wrong...of course, anyone can get the easy calls right, so it makes you wonder if they could be anymore worthless. The irony is, players throw helmets all the time...Braun was ejected because he threw his helmet and the embarrassment to the game probably knew he got it wrong, so he ejected him. Remember, good umpires hardly ever get noticed, so this fella, by blowing a call and then throwing a player out, pretty much proved, in two steps, that he is not capable of being on the field. One of the neighbors just carried the Rambling dog back home...he was sniffing around her deck, she said. I checked his chain and his collar...both seem fine. Not sure what happened. Yesterday, Discovery had on some investigation shows in which they studied JFK's assassination (in which they recreated the "magic bullet" perfectly), Bigfoot sightings, and UFO sightings, and to sum up as quickly as possible, found that those nuts that claim to have seen stuff were either making bad at home movies or saw reflections and the like. Hence, instead of blaming an alien or ghost, I think I'll just assume something released when he stretched it tight. It's a good thing Parra has all the talent he has, or he'd simply be too frustrating to keep around. He's having the same issues he had before his demotion, laying his fastball right down the middle, making bad throws, letting a 1 or 2 run inning turn into a 3-4 run frame. Manny is capable of being a solid top-of-the-rotation guy, but sure can't put it together since the first half of last year. Mike Burns will probably pick up a couple more days on the big league roster...if the team was a couple games out of the race, they probably would have brought Chris Smith or someone else back from Nashville after last night's game, just to have another rested arm available. As is, with Suppan returning Tuesday and Bush Thursday, Burns is all but a sure thing to go out on one of those days, with Hernan Iribarren probably being the odd man out the other, as 13 pitchers is probably a safe bet with a pair of SP's who will be limited. A Brewers delegation made its annual visit to Walter Reed Army Medical Center to visit soldiers wounded in the wars in Iraq and Afghanistan. It's always a humbling experience, and Macha said general manager Doug Melvin came up with an idea to formally honor war heroes each season. "Doug said we honor Jackie Robinson every year for his contribution to the game and humanity," said Macha. "He said we should have a day every year when we honor the military throughout the majors. Everybody could wear camouflage hats." Macha said Melvin got the idea when he saw a camouflage jersey signed by reliever Scott Linebrink hanging in one of the wings. Linebrink formally played for the San Diego Padres, who wear camouflage jerseys at home every Sunday to honor the military. Rather than getting depressed, visitors to Walter Reed always have their spirits lifted by the spirit and positive attitudes of the wounded soldiers. "We visited 10 or 12 of them, and only one was down a little bit," Macha said. "They weren't feeling sorry for themselves. A lot of them wanted to go back (to Iraq and Afghanistan). That puts it all in perspective."--JS Of course, MLB does wear special caps on Memorial Day and July 4th (and I think 9/11), while as far as I know, no other sport does a damn thing. However, Doug is correct, a day dedicated to the military would be a fine idea. Heck, they already have camouflage caps, I saw a couple at the Miller Park store. The best part is, it would make the "hold hands and sing folk songs" crowd really mad, and that's plenty reason enough to get me in favor of anything. Freedom doesn't come cheap, it often is accompanied by death and suffering...yet no one ever argues it isn't worth the cost. Do you think Gamel will be able to produce enough as an everyday 3B next year? With the addition of Escobar, they're going to need offense from the other spots. Hank the Hammer First of all Hank, the Brewers' SS spot ranks 13th of 16 NL teams this year, and they are still 4th in runs scored, so I don't see Alcides bringing them down much, especially when you factor in his speed and baserunning making up some for his lack of assumed pop. Right now, the Crew ranks 8th of 16 at 3B OPS, higher than I would have guessed, with a line of .319/.414, 733 OPS. For comparison's sake, Gamel put up a .330/.434, 764 OPS when he started earlier this year, and his AAA numbers project to a major league equivalent of .320/.409, 729 OPS. With McGehee in the picture, any minor slump (a pair of 0-4's) would likely result in Mat getting a day off, so an extended time of struggle is unlikely. The C spot looks by far the easiest spot to pick up a few runs, as the Brewers have ranked 15th of 16 (though only 40 OPS points below the mean). UPDATE: Just a note, my proposed infield of Fielder/Weeks/Escobar/Gamel/McGehee/Counsell would cost about $18.5M, assuming Craig accepts $2M after a spectacular '09 campaign (I'm sure he could get more, but I hope he wants to stay in MIL). Considering Prince will make $10.5M by himself, that's a good price. --The Brewers have yet to place veterans like Mike Cameron and Trevor Hoffman on trade waivers, but with the team now out of contention, Rosenthal asks "why not?" --GM Doug Melvin is disinclined to make such deals because he knows the return may not be significant, but Hoffman in particular would be attractive. The Rockies have already discussed him internally. --Craig Counsell, Braden Looper, and Jason Kendall could have value as well.--Rosenthal, through MLBTR It seems obvious Doug is waiting to see if a team will claim him and then make an offer, after his team is out of the race. While it is unlikely, the Crew could win 9 out of 12, SL slump badly, and lose 9 out of 12, and suddenly, they're 3 or 4 games out. Just wondering if you were going to do a year end prospects list to go along with your preseason one. Other than missing Dykstra, which pretty much everyone did, you were pretty much right on. Also, was wondering who you thought might be next year's sleeper, both as a position player and pitcher, to compare to this year's McGehee and DiFelice? PJ Sleepers are tough, because Casey was not with the team until December. Going by who they have now, I'd guess Angel Salome and John Axford...though Axford is just a shot in the dark, as no one on the 40 man right now looks to be an option. Salome, a top prospect, isn't a classic sleeper either, but I think he'll contribute far more than most do. I can't see doing more than one list a year, as that seems to be overkill. Besides, why do one after the season, as a "new" one next March would be the same list. As for the list, Dykstra, Jeffress, and Brewer will all fall off, though an argument could still be made for Jeremy. Mat was serviceable while in the bigs, especially when he started, but really struggled when he went back to Nashville. However, if you take a look, he has recovered nicely, and overall, has put together a very good campaign. I was of the opinion that a Gamel/Escobar/Weeks/Fielder 2010 infield made a lot of sense, in a perfect world, with McGehee and Counsell in reserve (Casey and Craig each being used to "protect" the left side youngsters, though Gamel has done very well versus AAA LHP). I would probably still lean that way, but I doubt I would announce those were my plans, instead, choosing to divert as much pressure from Gamel as I could by saying "he'll be in the mix", or something along those lines. The big changes in the '10 version of the Crew's positional players will be in the OF and possibly at C. The big decision will be whether Cameron and Kendall return, and also, if they are, at what cost? The other impending question is what Hardy will bring back in trade...while we would love to see someone part with a young SP with upside for JJ, a pre-arby OF might be a more realistic option. Braun has stated he thinks RF is easier than LF, does the braintrust feel Hart can play CF...those are queries that will have to be answered as well, as if they need a guy that can play CF, the options are quite limited...while if you can plug the new guy into any OF spot, it opens up a ton of options, including the Adam Dunn types, offensive weapons that can only play a corner spot (preferably LF). Prince Fielder blasted a two-run homer in a win over the Nationals on Friday night. Fielder now has 33 dingers to go along with a major-league best 110 RBI. The big man is almost a sure-thing to beat his previous career-best of 119 RBI from 2007, but he could also pass Cecil Cooper for the franchise-record at 126.--Rotoworld Every once in a while, something ends up online that you shake your head at. He's got 110 RBI's in 121 games, and "he could" end up with 126? Barring injury, it would take a collapse not to. As you probably suspected, I am unhappy about the release of a convicted mass murderer, but moreso, I just keep wondering, isn't the world supposed to be head over heels in love with us now? Or, is there still envy because of our wealth and power? Sadly, despite being the world's relief agency (while the UN twiddles their thumbs and does nothing), spending billions just because it's the right thing to do, the US gets no credit for their charitable ways. The scariest thing about Social Security is this...no one knows how to get rid of it even if we wanted to...once a government program begins, it's pretty much on the books forever. When W tried to add choice to the program...you know, individuals choosing what to do with their own money, it was proudly defeated by the Congress, most of which are still there. No offense to Chris, who is certainly better than the 8 ERA he put up with the Brewers earlier this year, but this move points out how weak the AAA pitching depth is for the Crew. However, if you check out what Narveson has done at AAA in the link above, he has struck out more than a man per inning, and has a near 3-1 K/BB ratio...solid AAA stats, enough so where I almost find it hard to believe not a single team claimed him on waivers a while back. I'm surprised Chris Smith was not brought up, as he was useful, while far from dominant, in his couple month long stint in MIL. This may well have been a move to assist in keeping Adam Dunn and other DC lefty sluggers held down for the weekend, though Macha has not used Narveson or Swindle as a loogy when they have been in the majors. The Rambling brother e-mailed to ask about JJ Hardy's salary situation. It occurred to me that others may be wondering about this as well. Hardy is making $4.65M this year, and there is no sign of any sort of a "split" contract being utilized, so JJ is making that in Nashville as well. Often when on rehab, big leaguers have a post-game spread catered or delivered, as the cost of such an indulgence is minimal compared to their salary...no word if JJ is doing this or not.:) He is not under contract for '10, but is still under team control, regardless of his service time issues. The Crew can just part ways with Hardy by not offering him arby, which will make him a free agent (this is how the Red Sox acquired Dave Ortiz); they can resign him after non-tendering him for any amount; or they can offer him arby, which is the only way to not allow 29 other teams to talk to him. If he were to be non-tendered, a 27 year-old who has had success in the majors, who plays SS well, would have a multitude of interest, to say the least. At minimum, he would likely sign a 1-2 year deal for $6-7M a year. If he were offered arby, the Crew can only decrease his '09 salary by 20%, meaning they could not go lower than about $3.7M. Realistically, he'd still get a raise, as his career stats are still very good for a SS, and his defense is still OK. I would have expected him to get about $8M had he had a "normal" year, so after an "off year", if you split the difference, that will still put him at about $5-6M. Luckily, if you're looking at JJ as a trade chip, that is not a really high amount, unless you feel he'll forever be stuck at his '09 production level. As I said to the Rambling brother, there really is very little debate as far as the Brewers are concerned...Escobar at $425K is by far a better option than Hardy at $5-6M...if you can argue that, either you have little understanding of financial value per production, or you simply want Mark A to throw away his money. Add into that you also will have what Hardy brings back in trade, and the decision is very clear...I just wish Hardy's value would be what it was a few months ago. This gent thinks Lopez will get 4y/$30M, which would mean the Crew would easily be able to offer Felipe arby and have no worries about him accepting. The draft picks would be better than any prospects traded for, not worrying about the money needed to sign them, of course. I promise you MA and Doug will try to win every game they can. However, the one thing I hope they sit down and discuss (if they haven't already, Doug flew out to DC Wednesday, so he may have met with Ken and the coaches today or will tomorrow morning) what players they will try to keep and which they will not. For instance, if they do not feel they can offer Lopez arby, fearful that he will accept, they should put him on waivers and make a deal, as he's too talented to lose for nothing. Ditto for Cameron, Looper, Counsell, and a couple others. The Rambling wife and I got back from a Rambling vacation last evening. I'm off until next Tuesday, so I'll try and get a recap up tomorrow. We saw the last 2 Brewers' wins, and enjoyed visits to five establishments that have been featured on Diners, Drive Ins and Dives. We rode a train at Wisconsin Dells, enjoyed the cool breeze off Lake Michigan, and enjoyed yummy seafood at Joe's Crab Shack. We did not spend one cent on a hotel room (thanks credit card points), and found out it's always rush hour near downtown Chicago. We found out Subway and Arby's run great deals in Michigan ($2 6" meatball and cold cut combo at Subway, or 2 for $4 subs at Arby's), but not at home. I'm not sure how many it will take before some decide the feds are the ones you least want to spend money effectively. What's that old saying, it will take a billion dollars for them to distribute a million dollars? BA and Rock wondering what Ronny Cedeno is doing differently...I would say he's finally just getting a chance to play every day...he was spectacular (.400/.500) in AAA, and may well just need someone to put him in the lineup and get him 500 AB's. Members of the U.S. 3rd Battalion, 75th Ranger Regiment closed space with the enemy, apparently killing at least ten. Corporal Benjamin Kopp was shot and evacuated to Germany, then back to the United States, where he died just over a week later on 18 July. Benjamin was 21 years old and at the very tip of the spear. If not for such men, we would be at the mercy of every demon. Benjamin Kopp and his comrades were delivering the latest bad news to the sort of people who harbored the terrorists who attack innocent people around the world every day, and who attacked us at home on 9/11. Ranger Kopp was a veteran with three combat tours. He knew the risks, yet continued to fight. Benjamin was laid to rest at Arlington National Cemetery. Secretary of Defense Robert Gates quietly attended the funeral, as did my good friend, Colonel Erik Kurilla, the new commander of Ranger Regiment, where Kopp served until America lost one of its finest Sons. Yet the effect of Corporal Kopp did not end on the battlefields of Afghanistan; he only regrouped and continued to serve. Corporal Kopp had volunteered as an organ donor and his heart was transplanted. Two days after most people would have died, Benjamin Kopp’s heart was transplanted into Judy Meikle. According to the Washington Post, Meikle said, "How can you have a better heart?" said a grateful Judy Meikle, 57, of Winnetka, Ill., who is still recovering from the surgery. "I have the heart of a 21-year-old Army Ranger war hero beating in me."--Michael Yon Where, I ask, do we find such brave men and women as Benjamin Kopp? Read the whole thing here. Yon, by the way, does a fantastic job of reporting on the battles that the old media has ignored since they made a fool of themselves declaring Iraq "lost" as it was being won, despite the claims of Hank Reid, Nancy Pelosi, and many other dimwits. Not only does he cover these conflicts for free, he gives an unbiased account and finds stories like this that NBC ignores, because it doesn't fit the Olbermann/Matthews "tingle in my leg" agenda. Yon is supported 100% through donations, and if I were you, I would not wait for Keith or Chris to hit the Paypal button to keep this outstanding work coming...after all, when the morning shows is bringing you "Jon and Kate" interviews, while Mike Yon is doing this, those that wonder why the networks are failing need look no further than content. Jack Z picks up Bill Hall for SEA, while the Crew picks up a too-old hard thrower whose had success at low A ball, while apparently getting SEA to pick up a portion of his '10 salary...I'll guess about $1M. Adam reports Doug thinks he may be able to trade Hall. With the Reds known to like Bill, and Harang and Arroyo both have similar size deals (actually owed $2-4M more) next year, that seems far too easy to ever actually happen. Who knows who else might like Hall, the Yankees were known to have interest last offseason. UPDATE: Just to make clear, a Hall for Harang swap seems ideal for the Brewers, but even picking up $3M extra, Harang is far better, as Hall hasn't hit RHP for years. Seems likely the Crew would have to pick up cash or send an extra player. He's probably right, but he's way too optimistic. Things can turn around in a week, never mind 1.25 years. The reason Jimmy Carter is remembered as a complete failure is he started off slow and went down hill from there. Obama has plenty of time to make himself a media darling again. I'm not going to spend a lot of time on it right now, but you gotta wonder if Doug will become a "seller", as the Crew is 2 games under .500, 8.5 behind SL, and 7 behind in the wildcard. Those numbers would certainly indicate a team not in the race. Cameron, Hoffman, Lopez, and Looper would be the guys who would bring the most, but none would be dealt if they are in the long-term plans...which in this case, would be 2010. There are several role players, including Counsell, Catalanotto, Kendall, and Gerut, that might interest a contender as well. The Crew has now signed their top 20 draft picks (including Scooter Gennett, who has signed now as well...he would have been a 3rd/4th round choice had he not said he was going to college), including several that fell due to signability issues for well over slot money. They needed to rebuild the system after dealing over a dozen prospects in the past couple years. Adding a bunch of extra youngsters can't hurt. The Brewers had their sixth home-run replay of the season and third at Miller Park when Pence hit what was called a homer to left. The ball appeared to cross in front of the foul pole and land in the second deck. The umpires agreed and called it foul after looking at a replay.--JS That's 0-6 for the year. No tag plays, no watching the foot and listening for the ball, no imaginary lines crossed with a check swing or not. This is, as easy can be "what side of the pole did the ball go past on?" stuff. My dog could do it, if he had the ability to reason and lift up the proper paw. This is the state of major league umpiring. The days of even pretending they are anything but horrendous is long past. Consistently wrong is the only description that is accurate. Tom H is reporting the Crew likely has a deal with supplemental 1st round pick Kentrail Davis, but won't announce it 'til tomorrow. I have read elsewhere they have signed 4th rounder Bo Hall, well above slot, and are awaiting MLB's approval. ---Lopez has just been outstanding, especially since he's obviously at less than 100%. Defense is not his strength, much like Weeks. I won't argue he is so good he can skate by a bit...just like Braun and Fielder, actually. I wish we had more guys like that, not less. Gamel will probably be a better player than Felipe soon, but isn't now. This was a move by Doug that reminds you of just how Melvin "gets it". --Speaking of gifted, have you seen Escobar play SS? Golly, he's smooth. He reminds me of a pair of things...one, how Rickie Weeks looked at the plate in A ball the year he was drafted...just above it all, like no one could get him out. Two, Dennis Miller's great line about how a black man just sort of begins moving with the music, while a white fella makes a conscious decision to begin dancing. Most shortstops are fine athletes who have been taught to play SS...Alcides looks born to do so. He may not hit a whole lot next year, but the way he runs, and seems willing to just slap the ball, he won't be hitting .220 either. The folks who were not extremely excited about this kid are going to have egg on their face soon...they probably already do. ---Burns pitched nicely last night, but got lucky too...Lee hit a couple rockets to Counsell, Cat made a nice grab in RF as well. Today, the problem was not a couple HR's, the issue was walks (and a HBP) before the homers. Solo HR's rarely beat you, but HR's after free baserunners will make it very tough to get a W. ---Braun and Fielder are both seeing the ball well. If they can keep it up for a week or so, it will go a long way in winning 8 of 10 to get back to the point contention is not "what if". This is probably the most unintentionally funny thing I have ever posted here. The stress the authors feel over flushing or not, and composting your own waste is so over the top...I'm still laughing and I read it 10 minutes ago. Enjoy. UPDATE: Glenn strikes again. The transparency makes her nothing but a caricature of herself. Seriously, when one just says the exact opposite from one moment to the next, I see them as ineffective clods. Know what? I'm for minimal government, low taxes, and a strong national defense. Next week, next month, 3 years from now, I'm still going to be for the same things. Pelosi just goes where the breeze takes her. If I'm a Democrat, I'm embarrassed and looking for new, stronger leadership. They might even sign a couple picks they did not plan on, guys drafted late that were expected to go to college, that were likely top 5 round talent had they been considered "signable". They might be using some money budgeted for adding a mythical pitcher that last place teams can't bear to deal. If Bourgeous takes over as the RF versus LHP, his splits over his career look pretty good. His walks when a southpaw pitches are especially good to see. As a speedy, all-around utility guy, he has a chance to impress and win a spot in '10. I'm shocked to see Hernan be honored for his defense, as the one time I saw him, he looked as if he could barely field a ground ball without stumbling. He's out of options, I like him as a 2B/OF reserve, and cheap lefty line drive bat for '10. I realize that myself giving AG.com hits for his fine link report is redundant, but it's a high point of my week, and I like to share. Sure that's kind of sad, but it's true. 8/14/2009 03:32:00 PM These are the good old days. Some folks are just too busy wishing the streets were paved with gold to enjoy the good times. Whatever strikes me as interesting, and serious Milwaukee Brewers thoughts. If you are a believer in respecting OBP, throwing strikes, and keeping the ball in the park, you may have found the place you've been searching for. I believe in low taxes, small government, and am not afraid to be labeled patriotic. If you are interested in sausage race results, walk up music, or professional wrestling, you may wish to click elsewhere. I'm happy to pay taxes to help the helpless. I don't like paying taxes to help the clueless. Look at the Occupy movement...I'm forced to pay taxes to help those whose plight I delight in.--Dennis Miller If you choose the path of terror, your life will be empty, and your life will be brief.--President Trump Never have lives less lived been more chronicled.--Dennis Miller I’m going to plead with you, do not cross us. Because if you do, the survivors will write about what we do here for 10,000 years.--Mad Dog Mattis I have never understood why it is "greed" to want to keep the money you have earned but not greed to want to take somebody else's money.--Thomas Sowell

[ { "description": "minLength validation", "schema": {"minLength": 2}, "tests": [ { "description": "longer is valid", "data": "foo", "valid": true }, { "description": "exact length is valid", "data": "fo", "valid": true }, { "description": "too short is invalid", "data": "f", "valid": false }, { "description": "ignores non-strings", "data": 1, "valid": true }, { "description": "one supplementary Unicode code point is not long enough", "data": "\uD83D\uDCA9", "valid": false } ] } ]

Next video in 15 Secs It's a common thinking for women to label 'All Men Are The Same'. But generalization aside, there are many gentlemen out there too! Neither are all men exploding hormones every minute, nor every Indian man is out there to rape someone. In short Not All Men Are Alike. Courtesy : Aisle

Search form Sign in with Twitter Sheena Queen of the Jungle # Review by Olivier Roth on March 28, 2018 Written by: Marguerite Bennett and Christina Trujillo Art by: Maria Sanapo Color: Ceci de la Cruz Letters by: Thomas Napolitano Published by: Dynamite The battle to stave up the poison that threatens to destroy the jungle that is home to Sheena and her tribe continues this month with a pretty surprising twist. At the end of the last issue, Sheena finally caught up to the Huntress, the supposed cause of the jungle’s sickness, and subsequently confronts her in a good old fashioned jungle fight. What happens next, and the revelation of the Huntress’s identity is the surprising twist. What happens after, not so much. Since most of the first half of the issue deals with Sheena’s confrontation with the Huntress and talking about any plot points would be super spoilery, I will concentrate on the second half of the issue - the evil corporation! So we learn that the evil Caldwell corporation (the bad guys from the first story arc) have gotten their hands on an artifact from the jungle that looks like a very large tomb thing. What Caldwell is unable to do is gain access to this tomb that they paid quite a handsome sum for. After losing a few teams in the attempt at opening the door, the head of Caldwell learns from one of his minions that there is a woman was able to close a door very similar to the one in their way, so she may be able to open it as well. Queue attempt at kidnapping Sheena! The issue itself is not bad, and never really has been, but it’s also a case that it feels like a lot of deja vu for anyone who has ever read any type of jungle pulp in the past. There is no real new territory that is touched upon within this issue, or this series, but every time I finish reading an issue of Sheena, I was have a glimmer of hope. The potential is there, it’s just not being fulfilled. Sanapo with the help of de la Cruz continues this month some great visuals in the comic itself. Her art skirts on the edge of the “bad girl” art of the 90s - mostly due to the subject matter more than by choice - but never fully goes over the line. I’m hoping to see some improvement in the expressions of her characters, something I’m a big proponent of, but otherwise, a pretty good job on her part.

Q: OAI field in biblatex/biber Similarly to DOI, will an OAI (Open Archives Initiative) field be soon added to bibtex/biblatex styles? A: With the help of experts, here is one solution that satisfies oai requirements. The oai references require a repository and a number such as, for instance oai:hal.example.com:hal-0000001 that can be used to reconstruct the actual internet address of the document of interest, here : http://hal.example.com/hal-000001/ The eprint and eprinttype fields together with the following definition shall be used: \def\oaitourl#1:#2{#1/#2} \DeclareFieldFormat{eprint:oai}% {oai:\href{http://\oaitourl#1}{\nolinkurl{#1}}} \begin{document} \cite{EX} \end{document} together with, in the .bib file : @TechReport{EX, author = {Name, First name}, title = {Report}, institution = {INH}, year = {2010}, eprint = {hal.example.com:hal-000000001}, eprinttype = {oai}} The thing is that it would be even better with an oai field directly in the .bib file. edit: This approach requires the use of the biblatex package together with the biber engine.

Rank Checker Tool Released by SEO Book Aaron Wall has written a pretty awesome rank checker tool Firefox plugin that checks various keywords against results on Google, Yahoo, and Microsoft Live. Here's a sample screenshot of what you're expected to see: While it looks like it's pretty accurate from my perspective, on Sphinn, one of the users note that it isn't completely accurate. Aaron has responded to inform us that the tool "clusters 100 results per page to minimize scraping on Google. That clustering can cause a shift in ranking compared to only seeing 10 results per SERP." But I gotta say -- this is probably one of the tools that should have been made earlier and just hasn't been. Now, you really don't need an API key to run ranking tools. You can do it out of the convenience of your web browser. Thanks, Aaron. This rocks.

--- abstract: | In the classical model for (information theoretically secure) Private Information Retrieval (PIR) due to Chor, Goldreich, Kushilevitz and Sudan, a user wishes to retrieve one bit of a database that is stored on a set of $n$ servers, in such a way that no individual server gains information about which bit the user is interested in. The aim is to design schemes that minimise the total communication between the user and the servers. More recently, there have been moves to consider more realistic models where the total storage of the set of servers, or the per server storage, should be minimised (possibly using techniques from distributed storage), and where the database is divided into $R$-bit records with $R>1$, and the user wishes to retrieve one record rather than one bit. When $R$ is large, downloads from the servers to the user dominate the communication complexity and so the aim is to minimise the total number of downloaded bits. Work of Shah, Rashmi and Ramchandran shows that at least $R+1$ bits must be downloaded from servers in the worst case, and provides PIR schemes meeting this bound. Sun and Jafar have considered the download cost of a scheme, defined as the ratio of the message length $R$ and the total number of bits downloaded. They determine the best asymptotic download cost of a PIR scheme (as $R\rightarrow\infty$) when a database of $k$ messages is stored by $n$ servers. This paper provides various bounds on the download complexity of a PIR scheme, generalising those of Shah et al. to the case when the number $n$ of servers is bounded, and providing links with classical techniques due to Chor et al. The paper also provides a range of constructions for PIR schemes that are either simpler or perform better than previously known schemes. These constructions include explicit schemes that achieve the best asymptotic download complexity of Sun and Jafar with significantly lower upload complexity, and general techniques for constructing a scheme with good worst case download complexity from a scheme with good download complexity on average. author: - 'Simon R. Blackburn[^1]' - 'Tuvi Etzion[^2]' - 'Maura B. Paterson[^3]' title: | PIR schemes with small download complexity\ and low storage requirements[^4] --- Introduction ============ The PIR Model ------------- In the classical model for private information retrieval (PIR) due to Chor, Goldreich, Kushilevitz and Sudan [@CGKS98], a database $\mathbf{X}$ is replicated across $n$ servers $S_1,S_2,\dotsc,S_n$. A user wishes to retrieve one bit of the database, so sends a query to each server and downloads their reply. The user should be able to deduce the bit from the servers’ replies. Moreover, no single server should gain any information on which bit the user wishes to retrieve (without collusion). The resulting protocol is known as an (information-theoretic) *PIR scheme*; there are also computational variants of the security model [@KuOs97]. The goal of PIR is to minimise the total communication between the user and the servers. In practice, the assumption that the user only wishes to retrieve one bit of the database, and the assumption that there is no shortage of server storage seem unrealistic. Because of this, many recent papers assume that the database $\mathbf{X}$ consists of $k$ [*records*]{}, each of which is $R$ bits in length, so that the number of possible databases is $2^{kR}$. We denote the value of Record $i$ by $X_i$, and we write $X_{ij}$ for the $j$th bit of $X_i$. The aim of the protocol is for the user to retrieve the whole of $X_i$, rather than a single bit. We also, following Shah, Rashmi and Ramchandran [@SRR14], drop the assumption that the whole database is replicated across the $n$ servers $S_1,S_2,\ldots,S_n$ and so, for example, there is the possibility of using techniques from coding theory in general and from distributed storage codes in particular to reduce the total storage of the scheme. No restrictions are made on the particular encoding used to distribute the database across the servers other than to assume it is deterministic, [*i.e.*]{} that there is a unique way to encode each database. This important generalisation of the model has led to very interesting recent work which we discuss in Subsection \[subsec:context\] below. More combinatorially, we define a private information retrieval scheme as follows. Suppose a database $\mathbf{X}$ is distributed across $n$ servers $S_1,S_2,\dotsc,S_n$. A user who wishes to learn the value $X_\ell$ of Record $\ell$ submits a [*query*]{} $(q_1,q_2,\dotsc,q_n)$. For each $r\in \{1,2,\dotsc, n\},$ server $S_r$ receives $q_r$ and responds with a value $c_r$ that depends on $q_r$ and on the information stored by $S_r$. The user receives the [*response*]{} $(c_1,c_2,\dotsc,c_n)$. This system is a [*private information retrieval (PIR) scheme*]{} if the following two properties are satisfied: **(Privacy)** For $r=1,2,\dotsc,n$ the value $q_r$ received by server $S_r$ reveals no information about which record is being sought. **(Correctness)** Given a response $(c_1,c_2,\dotsc,c_n)$ to a query $(q_1,q_2,\dotsc,q_n)$ for Record $\ell$, the user is unambiguously able to recover the value $X_\ell$ of this record. Note that while the query is drawn randomly according a pre-specified distribution on a set of potential queries, the response is assumed to be deterministic. \[ex:1server\] In the case of a single server, a trivial method for achieving PIR is for the user to download the entire $kR$-bit database. Chor, Goldreich, Kushilevitz and Sudan showed that in the case of single-bit records ($R=1$), if there is a single server then PIR is only possible if the total communication is at least $k$ bits ([*i.e.*]{} the size of the entire database) [@CGKS98], and so the solution above is best possible. We are interested in finding solutions such as the scheme below, which transmit significantly fewer than $kR$ bits. [@CGKS98] \[ex:chor\] Suppose there are two servers, each storing the entire database. Suppose $R=1$. - A user who requires Record $\ell$ chooses a $k$-bit string $(\alpha_1,\alpha_2,\dotsc, \alpha_k)$ uniformly at random. - Server 1 is requested to return the value $c_1=\bigoplus_{i=1}^k \alpha_i X_i$, and Server 2 is requested to return $ c_2=\left(\bigoplus_{i=1}^k \beta_i X_i\right)$, where $$\beta_i=\begin{cases} \alpha_i\oplus 1&\text{when }i=\ell,\\ \alpha_i&\text{otherwise.} \end{cases}$$ - The user computes $c_1\oplus c_2$ to recover the value $X_\ell$ of Record $\ell$. The strings $(\alpha_1,\alpha_2,\dotsc, \alpha_k)$ and $(\beta_1,\beta_2,\dotsc, \beta_{k})$ are both uniformly distributed, and are independent of the choice of $\ell$, hence neither server receives any information as to which record is being recovered by the user. We note that the scheme above works unchanged when the records are $R$-bit strings rather than single bits. The download complexity of the scheme, in other words the total number of bits downloaded from the servers, is $2R$. The upload complexity is $2k$, since each server receives a $k$-bit string from the user. Thus the total communication of the scheme is $2R+2k$ bits, which is significantly less than $kR$ bits for most parameters. Note that the upload complexity of this scheme does not depend on $R$, and so is an insignificant proportion of the total communication when $R$ is large. This is a general phenomenon: Chan, Ho and Yamamoto [@CHY15 Remark 2] observe the following. Let $m>1$ be an integer. Suppose we have an $n$-server PIR scheme for a database of $k$ records, each $R$ bits long. Suppose the scheme requires $u$ upload bits and $d$ download bits. Then we can construct an $n$-server PIR scheme for a database of $k$ records, each $mR$ bits long, which requires $md$ bits of download but still needs just $u$ bits to be uploaded. Note that when $m$ is large (so records are long) the communication complexity of the new scheme is dominated by the download complexity of the given scheme. Because of the observation of Chan et al., it is vital to find PIR schemes with low download complexity. We formalise the download complexity as follows. A PIR scheme *uses binary channels* if the response $c_j$ sent by server $S_j$ is a binary string of length $d_j$, where $d_j$ depends only on the query $q_j$ it receives. The *download complexity* is the maximum of the sum $\sum_{j=1}^n d_j$ over all possible queries $(q_1,q_2,\dotsc,q_n)$. So the download complexity is the number of bits downloaded in the worst case. We emphasise that the length $d_j$ in the definition above does not depend on the database $\mathbf{X}$, but could depend on the query $q_j$ received by server $S_j$. We note that we allow for the possibility that $d_j=0$, so the server does not reply to the query. Finally, we note that if we know that there are more than $2^x$ distinct possibilities for $c_j$ as the database varies, we may deduce that $d_j\geq x+1$. Although it is possible to use non-binary channels for PIR, we restrict our exposition to PIR schemes using binary channels in this paper (as is standard in the literature). We should comment that, despite the observation of Chan et al., we should not ignore upload complexity completely, as there are scenarios (for example, when $R$ is not so large) when it might be dominant. Moreover, we cannot compare $2k$ bits of download with $2R$ bits of upload just by comparing $k$ and $R$, since we know (at least currently) that in practice it takes much more time to upload a bit than to download one. Of course, we don’t know how the speed of downloading and uploading will change over time. But the obvious consequence of the current situation and future developments is to consider both upload and download complexities separately, and not to ignore one of them completely. This is something that will be done in this paper, although the download complexity will be the main target for optimisation since we generally assume that the size of the database is (considerably) larger than the one bit of the classical PIR model. We continue to another important measure that has motivated many papers in the last three years, after being introduced by Shah et el.[@SRR14]: Suppose server $S_r$ stores $s_r$ bits of information about the database $\mathbf{X}$. - The *per-server storage* of the scheme is $\max\{ s_r\mid r=1,2,\ldots ,n\}$. - The *total storage* of the scheme is $\sum_{r=1}^ns_r$. - The *storage overhead* of the scheme is the ratio between the total storage and the size of the database, i.e. $kR$. The classical model of PIR ignored storage issues: it was assumed that there is enough storage to allow the replication of the database at each server. But, with the quantity of information stored today in data centres, storage is an issue today and might be an important barrier in the future. Thus, it is important to reduce the storage overhead as much as possible, while keeping reliability, fast access, fast upload and fast download at reasonable levels. This is the perspective of the current paper, which is concerned with schemes whose download complexity is as small as possible whilst keeping the total storage at reasonable levels. Our contributions ----------------- In Section \[sec:lower\_bounds\], we provide combinatorial results on the structure of a PIR scheme with small download complexity: - We generalise (Theorem \[thm:download\]) a key theorem in the foundational paper of Chor et al. [@CGKS98], and use this result to generalise the lower bound of $R+1$ on download complexity in [@SRR14]. The results imply (Theorem \[thm:tdownload\]) that an $n$-server PIR scheme must have download complexity at least $\frac{n}{n-1}R$ when $k>\lceil R/(n-1)\rceil$. (This last result can also be obtained as a corollary of a recent bound due to Sun and Jafar [@SuJa16b].) These results provide a bridge between classical PIR and the new models that are assuming the retrieval of long records. - We provide (Corollary \[cor:onebit\], Theorem \[thm:almostall\]) information on the structure of a PIR scheme with minimal download complexity $R+1$. In particular, Theorem \[thm:almostall\] provides a rigorous statement of [@SRR14 Theorem 1]. In Section \[sec:constructions\], we provide various constructions for PIR schemes with low download complexity: - In Subsection \[subsec:rplusone\], we provide two simple $(R+1)$-server PIR schemes with download complexity $R+1$. Both schemes have total storage which is quadratic in $R$. The first scheme is a natural generalisation of the scheme of Chor et al. given above. The second scheme is a close variant of the quadratic total storage PIR scheme in [@SRR14], which avoids having to design slightly different schemes depending on the parity of $R$. This second scheme is to be preferred due to its lower upload complexity. (Another, more complex, PIR scheme with download complexity $R+1$ is considered in detail in [@SRR14]. This scheme has small per-server storage, but requires an exponential (in $R$) number of servers, and so has exponential total storage.) - In Subsection \[subsec:nserver\], we describe an $n$-server PIR scheme with download complexity $\frac{n}{n-1}R$. The total storage of the scheme is linear in $R$. This shows that for any $\epsilon>0$ there exists a PIR scheme with linear total storage and download complexity at most $(1+\epsilon)R$. (Schemes with linear total storage, but with download complexity between $2R$ and $4R$, are given in [@SRR14].) - We describe (Subsection \[subsec:perserver\]) schemes that provide trade-offs between increasing the number of servers and reducing the per-server storage of the scheme in Subsection \[subsec:nserver\]. - In Subsection \[subsec:sunjafar\], we provide explicit schemes that achieve optimal asymptotic download cost. The performance of these schemes is equal to the inductively defined schemes in Sun and Jafar [@SuJa16b], but the description of these schemes is more concise, and the proof that they are indeed PIR schemes is much more straightforward. - Finally, in Subsection \[subsec:averaging\], we explain an averaging technique that allows a PIR scheme with good average download complexity to be transformed into a scheme with good download complexity in the worst case. Context {#subsec:context} ------- We end this introduction with a discussion of some of the related literature. (Many of these papers appeared after the conference version of our paper [@BEM17] was posted.) Private information retrieval was introduced in [@CGKS98], and has been an active area ever since. See, for example, Yekhanin [@Y10] for a fairly recent survey. The papers by Shah et al. [@SRR14] and (independently) by Augot, Levy-Dit-Vahel, and Shikfa [@ALS14] are the first to consider PIR models where the information stored in the servers could be coded using techniques from distributed storage. Whereas [@SRR14] is mainly concerned with download complexity, and also with total storage (with per-server storage, and query size also relevant parameters), the authors of [@ALS14] emphasise measures of robustness against malicious servers, namely decoder locality and PIR locality. More recently, the literature has addressed several parallel and related issues, which can be categorised as follows: 1. Papers dealing with the download complexity, rate, and capacity of PIR schemes. 2. Research which attempts to reduce the storage overhead of PIR schemes. 3. Papers which present coding techniques, based on various error-correcting codes, e.g. MDS codes, to store the database in a distributed fashion. 4. Papers which consider PIR schemes in the presence of unreliable servers. Servers might be colluding (so they have access to more than one query $q_r$), they might fail (and so do not reply with a value $c_r$), they might be adversarial (replying with incorrect values $c_r$), they might be unsynchronised (storing slightly different copies of the database) and so on. 5. Research which aims to build PIR schemes into previously known architectures for distributed storage. 6. Papers dealing with other PIR models, for example allowing broadcasting of some information, or allowing the user to possess side information such as the value of some records. Clearly, these issues are related, and a given paper might address aspects of more than one of these topics. In early papers, Fanti and Ramchandran [@FaRa14; @FaRa15] considered unsynchronized databases; the results are the same as for synchronized PIR at the expense of probabilistic success for information retrieval, and the use of two rounds of communication. We are not aware of recent work in this model, but we mention in this context the work of Tajeddine and El Rouayheb [@TaElR17] which considers PIR schemes in the presence of some servers which do respond to a query. In a sequence of papers, Sun and Jafar [@SuJa16b; @SuJa16c; @SuJa16d; @SuJa16e; @SuJa16f] consider the capacity of the channels related to PIR codes in various models. (The *rate* of a PIR scheme is the ratio of $R$ and the download complexity, and the *capacity* is the supremum of achievable rates.) In the model for PIR we consdier, they use information theoretic techniques to show [@SuJa16b] that an $n$-server PIR scheme on a $k$ message database has rate at most $$\left(1-\frac{1}{n}\right)\left(1-\left(\frac{1}{n}\right)^k\right)^{-1}.$$ (Their model is restricted to the special case of replication, but it is easy to see that this restriction is not needed for this result to hold.) They also provide a scheme that attains this rate. The messages in their scheme are extremely long for most values of $n$ and $k$: the message length must be a multiple of $n^k$. Because of this, the scheme can be thought of as being tailored for the situation when $R\rightarrow\infty$. Their results show that when $R\rightarrow\infty$ with $n$ and $k$ fixed, there are schemes whose download complexity (and so whose communication complexity) has a leading term of the form $$\frac{n}{n-1}\left(1-\left(\frac{1}{n}\right)^k\right)R,$$ and that this term is best possible. (We give an explicit scheme with the same download complexity in Subsection \[subsec:sunjafar\].) The results in [@SuJa16b] have been generalised to the case when some of the servers collude. Sun and Jafar [@SuJa16c] find the capacity of the channel in this more general case. (The results in [@SuJa16b] can be thought of as the special case where each server can collude only with itself.) The capacity for the symmetric PIR model, where the user who retrieves a message will get no information about the other messages in the database, is determined in [@SuJa16d]. The optimal download complexity in the situation when the messages in the database might be of different lengths is considered in [@SuJa16e]. The most recent in this sequence of papers considers an interactive model, where a user can have several rounds of queries, the queries in a given round are allowed to depend on answers from previous rounds. Moreover, colluding servers are considered in this model. It is proved [@SuJa16f] that for this case there is no change in the capacity, but that the storage overhead can sometimes be improved. Banawan and Ulukus [@BaUl16] also generalise the results of Sun and Jafar [@SuJa16b], finding the capacity of the PIR scheme when the database is encoded with a linear code. Another generalisation due to Banawan and Ulukus [@BaUl17; @BaUl17a] is to the scenario that the user is allowed to request a few records in one round of queries. They provide capacity computations and schemes for this scenario. A similar case was also discussed in [@ZhXu17]. Finally, Banawan and Ulukus [@BaUl17b] consider the capacity of PIR schemes in the scenario where servers might not be synchronised, there might be adversarial errors, and some servers might collude. They compute the capacity when some or all of these events might occur. Wang and Skoglund [@WaSk16] consider the capacity of a symmetric PIR scheme when the database is stored in a distributed fashion using an MDS code. Chan, Ho, and Yamamoto [@CHY14; @CHY15] consider the trade-off between the total storage and the download complexity when the size of a record is large; the trade-off depends on the number of records in the system. They also consider the case where the database is encoded with an MDS code. Fazeli, Vardy, and Yaakobi [@FVY15; @FVY15a] give a method to reduce the storage overhead based on any known PIR scheme which uses replication. Their method reduces the storage overhead considerably, without affecting the order of the download complexity or upload complexity of the overall scheme, by simulating the original scheme on a larger number of servers. Their key concept is an object they call a $\kappa$-PIR code (more generally a $\kappa$-PIR array code), where $\kappa$ is the number of servers used in the originally known PIR scheme, which controls how a database can be divided into parts and encoded within servers to allow a trade-off between the number of servers and the storage overhead. In particular, for all $\epsilon>0$, they show that there exist good schemes (in terms of communication requirements) where the amount of information stored in a server is bounded but the total storage is at most $(1+\epsilon)$ times the database size. Rao and Vardy [@RaVa16] study PIR codes further, establishing the asymptotic behavior of $\kappa$-PIR codes. Vajha, Ramkumar, and Kumar [@VRK17; @VRK17a] find the redundancy of such codes for $\kappa =3,~4$ by using Reed-Muller codes. Lin and Rosnes [@LiRo17] show how to shorten and lengthen PIR codes, and find the redundancy of such codes for $\kappa =5,~6$. Blackburn and Etzion [@BE16; @BlEt17] consider the optimal ratios between $\kappa$-PIR array codes and the actual number of servers used in the system. Zhang, Wang, Wei, and Ge [@ZWWG16] consider these ratios further, and improve some of the results from [@BE16; @BlEt17]. We remark that though it is possible to reduce the storage overhead using the techniques of PIR array codes, it seems impossible to reduce the download complexity of the resulting schemes below $(3/2)R$ (and most codes give download complexity close to $2R$) because of restrictions on the PIR rate of such codes. It is interesting to note that Augot, Levy-Dit-Vahel, and Shikfa [@ALS14] constructed PIR schemes by partitioning the database into smaller parts, as done later in [@FVY15; @FVY15a], to reduce the storage overhead. But they applied this technique only to a certain family of multiplicity codes, and the parts of the partition were not encoded as in [@FVY15; @FVY15a]. Fazeli, Vardy, and Yaakobi [@FVY15a] remark that the concept of a $\kappa$-PIR code is closely related to codes with locality and availability. Such codes were studied first by Rawat, Papailiopoulos, Dimakis, and Vishwanath [@RPDV14b; @RPDV16] and later also by others, for example [@FFGW17; @HYUP15]. A new subspace approach for such codes was given recently in [@SES17; @SES17a]. Another family of related codes with similar properties are batch codes, which were first defined by Ishai, Kushilevitz, Ostrovsky, and Sahai [@IKOS04] and were recently studied by many others, for example [@AsYa17; @RSDG16]. It is important to note that all these codes are very important in the theory of distributed storage codes. This connection between the concepts of locality and PIR codes are explored in [@FFGW17]. Error-correcting codes, and in particular maximum distance separable (MDS) codes, have been considered by many authors in various PIR models. It is natural to consider MDS codes, as they are very often used in various types of distributed storage codes (especially for locally repairable codes [@GHSY12] and regenerating codes [@DGWWR10; @DRWS11]), and we expect that the servers in our PIR scheme will be part of a distributed storage system. We will now mention various examples. Colluding or malicious servers in PIR have been much studied over the last two years. Tajeddine and El Rouayheb [@TaElR16; @TaElR16a] consider PIR schemes where the information is stored using MDS codes. They describe schemes which have optimal download complexity in this model, as they attain the bounds in [@CHY14; @CHY15], in the situation when one or two ‘spies’ (colluding and/or malicious servers) are present. In the case of one spy (no collusion) a generalisation to any linear code with rate greater than half was given in [@KRA16]. Freij-Hollanti, Gnilke, Hollanti, and Karpuk [@FGHK16] give a PIR scheme coded with an MDS code which can be adjusted (by varying the rate of the MDS code) to combat against larger numbers of colluding servers. This scheme also attains the asymptotic bound on the related capacity of such a PIR scheme. This idea is generalised in [@TGKFHR17]. The results in the latter paper are analysed (and one conjecture disproved) by Sun and Jafar [@SuJa17; @SuJa17a]. Another scheme based on MDS codes which can combat large number of colluding servers is given by Zhang and Ge [@ZhGe17a]. A generalisation to the case where the user wants to retrieve several files is given by the same authors in [@ZhGe17b]. Wang and Skoglund [@WaSk17a] consider a symmetric PIR scheme using an MDS code, in which the user can retrieve the information about the file he wants, but can gain no information about the other files. This scheme attains the bound on the capacity which they derive earlier in [@WaSk16]. They have extended their work to accommodate colluding servers in [@WaSk17b]. PIR can be combined with other applications in storage and communication in many ways. One example is a related broadcasting scheme in [@KSCF17]. Another example is cache-aided PIR, considered by Tandon [@Tan17]. In this setup the user is equipped with a local cache which is formed from an arbitrary function on the whole set of messages, and this local cache is known to the servers. The situation when this cache is not known to the servers is considered by Kadhe, Garcia, Heidarzadeh, El Rouayheb, and Sprintson [@KGHRS17]. Since the user has side information in these models, the problem is closely related to index coding [@BBJK11] a topic which is also of great interest. Optimal download complexity {#sec:lower_bounds} =========================== In this section, we give structural results for PIR schemes with optimal download complexity, given that the database consists of $k$ records of length $R$. For some of the results, we also assume that the PIR scheme involves $n$ servers, where $n$ is fixed. In Subsection \[subsec:lower\_bounds\] we generalise a classical result in Private Information Retrieval due to Chor et al. We use this result to provide an alternative proof of the theorem of Shah, Rashmi and Ramchandran [@SRR14] that a PIR scheme must have download complexity at least $R+1$ when $k\geq 2$, and to prove a lower bound of $\frac{n}{n-1}R$ for the download complexity of an $n$-server PIR scheme whenever $k$ is sufficiently large. In Section \[subsec:R\_plus\_one\] we present more precise structural results when the download complexity of a PIR scheme attains the optimal value of $R+1$ bits. We say that a response $(c_1,c_2,\dotsc,c_n)$ is [*possible*]{} for a query $(q_1,q_2,\dotsc,q_n)$ if there exists a database $\mathbf{X}$ for which $(c_1,c_2,\dotsc,c_n)$ is returned as the response to the query $(q_1,q_2,\dotsc,q_n)$ when $\mathbf{X}$ is stored by the servers. Lower bounds on the download complexity {#subsec:lower_bounds} --------------------------------------- We aim to generalise the following theorem, which was proved by Chor et al. in the very first paper on PIR [@CGKS98 Theorem 5.1]: \[thm:chor\] A PIR scheme that uses a single server for a database with $k$ records of size one bit is not possible unless the number of possible responses from the server to any given query is at least $2^k$. Our generalisation shows a server must reply with at least $k(R-d)$ bits of download, if no more than at total of $d$ bits (where $0\leq d \leq R$) are downloaded from the other servers. We state our generalisation as follows. Without loss of generality we will focus on server $S_1$, so for ease of notation we will denote the tuple $(q_1,q_2,\dotsc,q_n)$ by $(q_1,q_{\ensuremath{\mathrm{other}}})$, and $(c_1,c_2,\dotsc, c_n)$ by $(c_1,c_{\ensuremath{\mathrm{other}}})$. \[thm:download\] Suppose $0\leq d \leq R$. Let $q_1$ be fixed. Suppose we have a PIR scheme with the property that for any query of the form $(q_1,q_{\ensuremath{\mathrm{other}}})$, we have $$|\{c_{\ensuremath{\mathrm{other}}}\mid\text{$\exists c_1$ such that $(c_1,c_{\ensuremath{\mathrm{other}}})$ is possible for $(q_1,q_{\ensuremath{\mathrm{other}}})$}\}|\leq 2^{d}.$$ Then for any query $(q_1,q'_{\ensuremath{\mathrm{other}}})$ we have $$|\{c_1\mid\text{$\exists c_{\ensuremath{\mathrm{other}}}$ such that $(c_1,c_{\ensuremath{\mathrm{other}}})$ is possible for $(q_1,q'_{\ensuremath{\mathrm{other}}})$}\}|\geq 2^{k(R-d)}.$$ We remark that Theorem \[thm:chor\] is the case $d=0$ and $R=1$ of Theorem \[thm:download\]. Let $q_1$ be fixed, and suppose we have a PIR scheme with the property that for any query $(q_1,q_{\ensuremath{\mathrm{other}}})$ $$|\{c_{\ensuremath{\mathrm{other}}}\mid\text{$\exists c_1$ such that $(c_1,c_{\ensuremath{\mathrm{other}}})$ is possible for $(q_1,q_{\ensuremath{\mathrm{other}}})$}\}|\leq 2^{d}.$$ Assume, for a contradiction, that there exists a query $(q_1,q_{\ensuremath{\mathrm{other}}}^\ast)$ corresponding to the $i$th record for which $$|\{c_1\mid\text{$\exists c_{\ensuremath{\mathrm{other}}}$ such that $(c_1,c_{\ensuremath{\mathrm{other}}})$ is possible for $(q_1,q_{\ensuremath{\mathrm{other}}}^\ast)$}\}|<2^{k(R-d)}.$$ There are $2^{kR}$ databases, and less than $2^{k(R-d)}$ possibilities for the reply $c_1$ of $S_1$ to the $(q_1,q_{\ensuremath{\mathrm{other}}}^\ast)$. So by the pigeonhole principle, there is a value $c_1^\ast$ for which there exists a set $T$ of databases with $|T|>2^{kR}/2^{k(R-d)}=2^{kd}$ having the property that for each $\mathbf{X}\in T$, the response of $S_1$ to $(q_1,q_{\ensuremath{\mathrm{other}}}^\ast)$ when the servers store $\mathbf{X}$ is $c_1^\ast$. If server $S_1$ receives the query $q_1$, it will thus return $c_1^\ast$ whenever a database in $T$ is being stored. Since the databases consist of $k$ records, the fact that $|T|>2^{kd}$ implies the existence of a record, say Record $\ell$, for which the number of distinct values $X_\ell$ that appear among the databases in $T$ is greater than $2^d$. Thus we can choose a set of databases $W\subseteq T$ such that no two databases in $W$ have the same value $X_\ell$. The requirement for privacy against server $S_1$ implies that if $(q_1,q_{\ensuremath{\mathrm{other}}}^\ast)$ is a query for Record $i$, then there exists a query for Record $\ell$ of the form $(q_1,q_{\ensuremath{\mathrm{other}}}^\ell)$, since otherwise $S_1$ could distinguish between queries for Record $i$ and Record $\ell$. If query $(q_1,q_{\ensuremath{\mathrm{other}}}^\ell)$ is made when a database in $T$ is stored, then server $S_1$ receives $q_1$ and responds $c_1^\ast$ as before. Now consider the databases in $W$. As there are $2^d+1$ of them, yet at most $2^d$ values for $c_{\ensuremath{\mathrm{other}}}$ for which there is a possible response $(c_1^\ast,c_{\ensuremath{\mathrm{other}}})$ to $(q_1,q_{\ensuremath{\mathrm{other}}}^\ell)$, it follows that there must be some value $c_{\ensuremath{\mathrm{other}}}^\ell$ for which there are two databases $\mathbf{X},\mathbf{Y}\in W$ such that the response to $(q_1,q_{\ensuremath{\mathrm{other}}}^\ell)$ is $(c_1^\ast,c_{\ensuremath{\mathrm{other}}}^\ell)$ when either of those databases is stored. This contradicts the correctness of our scheme, since the values of Record $\ell$ in $\mathbf{X}$ and $\mathbf{Y}$ are not equal yet the response $(c_1^\ast,c_{\ensuremath{\mathrm{other}}}^\ell)$ to the query $(q_1,q_{\ensuremath{\mathrm{other}}}^\ell)$ does not allow the user to distinguish between them. The following theorem is a key consequence of Theorem \[thm:download\]. \[thm:binarydownload\] Let $x$ be non-negative, and suppose we have a PIR scheme that has download complexity at most $R+x$. If the database contains $k$ records, where $k\geq x+2$, then the number of bits downloaded from any server is at most $x$. Without loss of generality, consider the server $S_1$. Suppose for a contradiction that there exists a query $q_1$ so that at least $x+1$ bits are downloaded from $S_1$ (and so at most $(R+x)-(x+1)=R-1$ bits are downloaded from the other servers). Suppose that a total of $d$ bits are downloaded from the other servers in the worst case when $S_1$ receives $q_1$. So $d\leq R-1$. Theorem \[thm:download\] implies that at least $k(R-d)$ bits are downloaded from $S_1$, and so at least $k(R-d)+d$ bits are downloaded from the servers in the worst case. But $d\leq R-1$ and $k\geq x+2$, so $$k(R-d)+d=kR-(k-1)d \geq kR-(k-1)(R-1)=R+k-1\geq R+(x+2)-1 =R+x+1,$$ which is impossible as the scheme has total download complexity $R+x$. This contradiction establishes the theorem. We are now in a position to provide a new short proof of the following corollary. The corollary is due to Shah et al. [@SRR14]. \[cor:rplusone\] Let the database contain $k$ records with $k\geq 2$. Any PIR scheme requires a total download of at least $R+1$ bits. Suppose we have a scheme with total download of $R$ or fewer bits. Theorem \[thm:binarydownload\] with $x=0$ implies that $0$ bits are downloaded from each server, and so the user receives no information about the desired record. Hence such a scheme cannot exist. The following theorem improves the bound of Corollary \[cor:rplusone\] when $n< R+1$ and $k$ is sufficiently large. \[thm:tdownload\] Suppose a PIR scheme involves $n$ servers, where $n\geq 2$. Suppose the database contains $k$ records, where $k\geq \lceil \frac{1}{n-1}R\rceil+1$. Then the download complexity of the scheme is at least $\frac{n}{n-1}R$ bits. Assume for a contradiction that the scheme has download complexity $R+x$, where $x$ is an integer such that $x<\frac{1}{n-1}R$. Since $x\leq \lceil \frac{1}{n-1}R\rceil -1$, we see that $k\geq x+2$ and so Theorem \[thm:binarydownload\] implies that the number of bits downloaded by any server is at most $x$. Since we have $n$ servers, the total number of bits of download is always at most $xn$. Since our scheme has download complexity $R+x$, there is a query where a total of $R+x$ bits are downloaded from servers. Hence we must have that $nx\geq R+x$, which implies that $x\geq \frac{1}{n-1}R$. This contradiction establishes the result. Download complexity $R+1$ {#subsec:R_plus_one} ------------------------- The final two results of this section concentrate on the extreme case when the download complexity is exactly $R+1$. Recall that the download complexity is a worst case measure, so $R+1$ is an upper bound on the number of bits downloaded for any query, and there exists a query where $R+1$ bits are downloaded. \[cor:onebit\] Let the database contain $k$ records with $k\geq 3$. Any PIR scheme with a total download of exactly $R+1$ bits requires 1 bit to be downloaded from each of $R$ or $R+1$ different servers in response to any query. The special case of Theorem \[thm:binarydownload\] when $x=1$ shows that no server replies with more than $1$ bit. For the download complexity to be $R+1$, no more than $R+1$ servers can respond non-trivially. Since the user deduces the value of an $R$-bit record from the bits it has downloaded, at least $R$ servers must reply to any query. One might hope that the Corollary \[cor:onebit\] could be strengthened to the statement that exactly $R+1$ servers must respond non-trivially. However, examples show that this is not always the case: see the comments after Construction \[con:manyserver\] below. Shah et al.  state [@SRR14 Theorem 1] that, in the situation above, “for almost every PIR operation” $R+1$ servers must respond, and they provide a heuristic argument to support this statement. The following result makes this rigorous, with a precise definition of ‘almost every’. \[thm:almostall\] Let the database contain $k$ records with $k\geq 3$. Suppose we have a PIR scheme with a total download of exactly $R+1$ bits (in the worst case). Suppose a user chooses to retrieve a record chosen with a uniform probability distribution on $\{1,2,\ldots,k\}$. Let $\alpha$ be the probability that only $R$ bits are downloaded. Then $$\alpha\leq \frac{R+1}{kR+1}.$$ By Corollary \[cor:onebit\], each server replies to any query with at most one bit. We may assume, without loss of generality, that if a server replies with one bit then this bit must depend on the database in some way (since otherwise we may modify the scheme so that this server does not reply and the probability $\alpha$ will increase). Let $(q_1,q_2,\ldots,q_n)$ be a query for the $\ell$th record where only $R$ servers reply non-trivially. Since only $R$ servers reply, there are at most $2^R$ possible replies to the query (over all databases). But the value $X_\ell$ of the record is determined by the reply, and there are $2^R$ possible values of $X_\ell$. So in fact there must be exactly $2^R$ possible replies, and there is a bijection between possible replies and possible values $X_\ell$. We claim that the replies of each of these $R$ servers can only depend on $X_\ell$, not on the rest of the database. To see this, suppose a server $S_r$ replies non-trivially, and let $f:\{0,1\}^{kR}\rightarrow\{0,1\}$ be the function mapping each possible value of the database to the reply of $S_r$ to query $q_r$. Suppose $f$ is not a function of $X_\ell$ alone, so there are two databases $\mathbf{X}$ and $\mathbf{X}'$ whose $\ell$th records are equal and such that $f(\mathbf{X})\not=f(\mathbf{X}')$. Let $\rho$ be the common value of the $\ell$th record in both $\mathbf{X}$ and $\mathbf{X}'$. When $X_\ell=\rho$ there are at least two possible replies to the query, depending on the value of the remainder of the database. But this contradicts the fact that we have a bijection between possible replies and possible values $X_\ell$. So our claim follows. Let $A$ be the event that exactly $R$ servers reply, and for $r=1,2,\ldots,n$ let $B_r$ be the event that server $S_r$ replies non-trivially. Let $D_r$ be the indicator random variable for the event $B_r$. So $D_r$ is equal to $1$ when $S_r$ responds non-trivially and $0$ otherwise. Note that $D_r$ is always equal to the number of bits downloaded from $S_r$, thus the expected value of the sum of these variables satisfies $$\label{eqn:deqn} {\mathrm{E}}\left(\sum_{r=1}^n D_r\right)=\alpha R + (1-\alpha)(R+1)=R+1-\alpha.$$ Let $D'_r$ be the indicator random variable for the event $A\wedge B_r$. When $A$ does not occur, all the variables $D'_r$ are equal to $0$. When $A$ occurs, $D'_r$ is the number of bits downloaded from server $S_r$ and a total of $R$ bits are downloaded. So $$\label{eqnddasheqn} {\mathrm{E}}\left(\sum_{r=1}^n D'_r\right)=(1-\alpha)0+\alpha R=\alpha R.$$ Suppose a server $S_r$ uses the following strategy to guess the value of $\ell$ from the query $q_r$ it receives. If the server replies non-trivially using a function $f$ that depends on only one record, say Record $\ell'$, it guesses that $\ell=\ell'$. Otherwise, the server guesses a value uniformly at random. The server guesses correctly with probability $1/k$ when it responds trivially. The argument in the paragraph above shows the server always guesses correctly if it responds non-trivially and only $R$ servers reply. Thus the server is correct with probability at least $(1/k)\Pr(\overline{B_r})+\Pr(A\wedge B_r)$. The privacy requirement of the PIR scheme implies that the server’s probability of success can be at most $1/k$, and so we must have that $\Pr(A\wedge B_r)\leq (1/k)\Pr(B_r)$. Hence $${\mathrm{E}}(D'_r)\leq (1/k){\mathrm{E}}(D_r).$$ By linearity of expectation, we see that $${\mathrm{E}}\left(\sum_{r=1}^n D'_r\right)=\sum_{r=1}^n{\mathrm{E}}(D'_r)\leq \frac{1}{k}\sum_{r=1}^n{\mathrm{E}}(D_r)=\frac{1}{k}\,{\mathrm{E}}\left(\sum_{r=1}^n D_r\right).$$ So, using  and , we see that $$\alpha R\leq \frac{1}{k}(R+1-\alpha).$$ Rearranging this inequality in terms of $\alpha$, we see that the theorem follows. Constructions {#sec:constructions} ============= Recall the notation from the introduction: we are assuming that our database $\mathbf{X}$ consists of $k$ records, each of $R$ bits, and we write $X_{ij}$ for the $j$th bit of the $i$th record. Two schemes with download complexity $R+1$ {#subsec:rplusone} ------------------------------------------ This section describes two schemes with download complexity $R+1$. Recall that this download complexity is optimal, by Corollary \[cor:rplusone\]. The first scheme is included because of its simplicity; it can be thought of as a variation of the scheme of Chor et al.  described in Example \[ex:chor\], and achieves optimal download complexity using only $R+1$ servers. It has a total storage requirement which is quadratic in $R$. But the scheme has high upload complexity: $kR(R+1)$. The second scheme is very closely related to a scheme mentioned in an aside in Shah et al. [@SRR14 Section IV]. This scheme has the same properties as the first scheme, except the upload complexity is improved to just $(R+1)k\lceil\log (R+1)\rceil$. We note that the main scheme described in Shah et al. [@SRR14 Section IV] also has optimal download complexity of $R+1$. Each server stores just $R$ bits, and so the storage per server is low. However, their scheme uses an exponential (in $R$) number of servers, and so has exponential total storage. \[con:manyserver\] Suppose there are $R+1$ servers, each storing the whole database. - A user who requires Record $\ell$ creates a $k\times R$ array of bits by drawing its entries $\alpha_{ij}$ uniformly and independently at random. - Server $S_{R+1}$ is requested to return the bit $c_{R+1}=\bigoplus_{i=1}^k\bigoplus_{j=1}^R\alpha_{ij}X_{ij}$. - For $r=1,2,\dotsc, R$, server $S_r$ is requested to return the bit $c_r=\bigoplus_{i=1}^k\bigoplus_{j=1}^R\beta_{ij}X_{ij}$, where $$\beta_{ij}=\begin{cases} \alpha_{ij}\oplus 1&\text{if }i=\ell\text{ and }j=r,\\ \alpha_{i,j}&\text{otherwise}. \end{cases}$$ - To recover $X_{\ell r}$, namely bit $r$ of record $X_\ell$, the user computes $c_r\oplus c_{R+1}$. \[thm:construction1\] Construction \[con:manyserver\] is a $(R+1)$-server PIR scheme with download complexity $R+1$. The scheme has upload complexity $kR(R+1)$ and total storage $(R+1)Rk$ bits. We note that $$\alpha_{ij}\oplus\beta_{ij}=\begin{cases} 1&\text{if }i=\ell\text{ and }j=r,\\ 0&\text{otherwise}. \end{cases}$$ Hence $$\begin{aligned} c_r\oplus c_{R+1}&=\bigoplus_{i=1}^k\bigoplus_{j=1}^R(\alpha_{ij}\oplus \beta_{ij})X_{ij}\\ &=X_{\ell r}.\end{aligned}$$ So the user recovers the bit $X_{\ell r}$ correctly for any $r$ with $1\leq r\leq R$. This proves correctness. For privacy, we note that $S_{R+1}$ receives a uniformly distributed vector $q_{R+1}=(\alpha_{ij})\in\{0,1\}^{kR}$ in all circumstances. Since the distribution of $q_{R+1}$ does not depend on $\ell$, no information about $\ell$ is received by $S_{R+1}$. Similarly, for any $1\leq r\leq R$, the query $q_r=(\beta_{ij})\in\{0,1\}^{kR}$ is uniformly distributed irrespective of the value of $\ell$, and so no information about $\ell$ is received by $S_r$. We note that each query $q_r$ is $kR$ bits long (for any $r\in\{1,2,\ldots,R+1\}$) and so the upload complexity of the scheme is $kR(R+1)$. Each server replies with a single bit, and so the download complexity is $R+1$. The database is $kR$ bits long, and so (since each server stores the whole database) the total storage is $(R+1)Rk$ bits. We note that there are situations where one of the servers is asked for an all-zero linear combination of bits from the database. In this case, that server need not reply. So the number of bits of downloaded in Construction \[con:manyserver\] is sometimes $R$ (though usually $R+1$ bits are downloaded). See the comment following Corollary \[cor:onebit\]. We now describe a second construction with improved upload complexity. The construction can be thought of as a variant of Construction \[con:manyserver\] where the rows of the array $\alpha$ are all taken from a restricted set $\{e_0,e_1,\ldots,e_{R}\}$ of size $R+1$. A similar idea is used in the constructions in [@SRR14]. For $i=1,2,\ldots,R$, let $e_i$ be the $i$^th^ unit vector of length $R$. Let $e_0$ be the all zero vector. For binary vectors $\mathbf{x}$ and $\mathbf{y}$ of length $R$, write $\mathbf{x}\cdot \mathbf{y}$ be their inner product; so $\mathbf{x}\cdot \mathbf{y}=\oplus_{j=1}^Rx_jy_j$. \[con:lowerupload\] Suppose there are $R+1$ servers, each storing the whole database. - A user who requires Record $\ell$ chooses $k$ elements $a_1,a_2,\ldots,a_k\in\mathbb{Z}_{R+1}$ uniformly and independently at random. For $r=1,\ldots,R+1$, server $S_r$ is sent the vector $q_r=(b_{1r},b_{2r},\ldots,b_{kr})\in\mathbb{Z}_{R+1}^k$, where $$b_{ir}=\begin{cases} a_i+r\bmod R+1&\text{if }i=\ell,\\ a_i&\text{otherwise}. \end{cases}$$ - Server $S_r$ returns the bit $c_{r}=\bigoplus_{i=1}^k e_{b_{ir}}\cdot X_i$. - To recover the $j$^th^ bit of $X_\ell$, the user finds the integers $r$ and $r'$ such that $b_{\ell r}=0$ and $b_{\ell r'}=j$. The user then computes $c_r\oplus c_{r'}$. \[thm:construction2\] Construction \[con:lowerupload\] is an $(R+1)$-server PIR scheme with download complexity $R+1$. The scheme has upload complexity $k(R+1)\log (R+1)$ and total storage $(R+1)Rk$ bits. For correctness, we first note that $r$ and $r'$ exist since $b_{\ell r}\in\{0,1,2\ldots,R\}$ takes on each possible value once as $r\in\{0,1,\ldots,R\}$ varies. Also note that $$e_{b_{ir}}\oplus e_{b_{ir'}}=\begin{cases} e_j&\text{if }i=\ell,\\ e_0&\text{otherwise}. \end{cases}$$ So, since $e_0=0$, $$c_r\oplus c_{r'}=\bigoplus_{i=1}^k(e_{b_{ir}}\oplus e_{b_{ir'}})\cdot X_{i}=e_j\cdot X_\ell =X_{\ell j}.$$ So the user recovers the bit $X_{\ell j}$ correctly for any $j$ with $1\leq j\leq R$. For privacy, we note that $S_r$ receives a uniformly distributed vector $q_{r}\in(\mathbb{Z}_{R+1})^{k}$ in all circumstances. Since the distribution of $q_{r}$ does not depend on $\ell$, no information about $\ell$ is received by $S_r$. The calculations of the total storage and download complexity are identical to those in the proof of Theorem \[thm:construction1\]. For the upload complexity, note that it takes just $\log (R+1)$ bits to specify an element of $\mathbb{Z}_{R+1}$. Since each server receives $k$ elements from $\mathbb{Z}_{R+1}$, and since there are $R+1$ servers, the upload complexity of the scheme is $k(R+1)\log (R+1)$ as claimed. Optimal download complexity for a small number of servers {#subsec:nserver} --------------------------------------------------------- For an integer $n$ such that $(n-1)\mid R$, we now describe an $n$-server PIR scheme with download complexity $\frac{n}{n-1}R$ bits. By Theorem \[thm:tdownload\], this construction provides schemes with an optimal download complexity for $n$ servers, provided the number $k$ of records is sufficiently large. This construction is closely related to Construction \[con:lowerupload\] above. Indeed, the construction below is a generalisation of Construction \[con:lowerupload\] where we work with strings rather than single bits. We first define an analogue of the bits $e_b\cdot X_i$ computed by servers in Construction \[con:lowerupload\]. We divide an $R$-bit string $X$ into $n-1$ blocks, each of size $R/(n-1)$. For $b\in\{1,2,\ldots,{n-1}\}$ we write $\pi_b(X)$ for the $b$th block (so $\pi_b(X)$ is an $R/(n-1)$-bit string). We write $\pi_0(X)$ for the all-zero string $0^{R/(n-1)}$ of length $R/(n-1)$. \[con:smallnumservers\] Let $n$ be an integer such that $(n-1)\mid R$. Suppose there are $n$ servers, each storing the entire database. - A user who requires Record $\ell$ chooses $k$ elements $a_1,a_2,\ldots,a_k\in\mathbb{Z}_{n}$ uniformly and independently at random. For $r=1,\ldots,n$, server $S_r$ is sent the vector $q_r=(b_{1r},b_{2r},\ldots,b_{kr})\in\mathbb{Z}_{n}^k$, where $$b_{ir}=\begin{cases} a_i+r\bmod n&\text{if }i=\ell,\\ a_i&\text{otherwise}. \end{cases}$$ - Server $S_r$ returns the $R/(n-1)$-bit string $c_{r}=\bigoplus_{i=1}^k \pi_{b_{ir}}(X_i)$. - To recover the $j$^th^ block of $X_\ell$, the user finds the integers $r$ and $r'$ such that $b_{\ell r}=0$ and $b_{\ell r'}=j$. The user then computes $c_r\oplus c_{r'}$. \[thm:construction3\] Construction \[con:smallnumservers\] is an $n$-server PIR scheme with download complexity $\frac{n}{n-1}R$. The scheme has upload complexity $nk\log n$ and total storage is $nkR$. Exactly as in the proof of Theorem \[thm:construction2\], we first note that $r$ and $r'$ exist since $b_{\ell r}\in\{0,1,2\ldots,n-1\}$ takes on each possible value once as $r\in\{0,1,\ldots,n\}$ varies. Also note that when $i\not=\ell$ $$\pi_{b_{ir}}(X_i)\oplus \pi_{b_{ir'}}(X_i)=\pi_{a_{i}}(X_i)\oplus \pi_{a_{i}}(X_i)=0^{R/(n-1)},$$ but when $i=\ell$ $$\pi_{b_{ir}}(X_i)\oplus \pi_{b_{ir'}}(X_i)=\pi_{0}(X_i)\oplus \pi_j(X_i)=\pi_j(X_i)= \pi_j(X_\ell).$$ Hence $$c_r\oplus c_{r'}=\bigoplus_{i=1}^k(\pi_{b_{ir}}(X_i)\oplus \pi_{b_{ir'}}(X_{i}))=\pi_j(X_{\ell}).$$ So the user recovers the $j$th block of $X_{\ell}$ correctly for any $j$ with $1\leq j\leq (n-1)$. For privacy, we note that $S_r$ receives a uniformly distributed vector $q_{r}\in(\mathbb{Z}_n)^{k}$ in all circumstances. Since the distribution of $q_{r}$ does not depend on $\ell$, no information about $\ell$ is received by $S_r$. The total storage is $nkR$, since each of $n$ servers stores the entire $kR$-bit database. Each query $q_r$ is $k\log n$ bits long, since an element of $\mathbb{Z}_n$ may be specified using $\log n$ bits. Hence the upload complexity is $nk\log n$. Since each server returns an $R/(n-1)$- bit string, the download complexity is $\frac{n}{n-1}R$. Shah et al. [@SRR14 Section V] provide PIR schemes with linear (in $R$) total storage and with download complexity between $2R$ and $4R$. Their scheme requires a number of servers which is independent of $R$ (but is linear in $k$). The construction above (taking $n$ to be fixed but sufficiently large) shows that for any fixed positive $\epsilon$ a PIR scheme with linear total storage exists with download complexity of $(1+\epsilon)R$: this is within an arbitrarily close factor of optimality. Moreover, the number of servers in our construction is independent of both $k$ and $R$. However, note that in our scheme each server stores the whole database, whereas the per server storage of the scheme of Shah et al.  is a fixed multiple of $R$. This issue is addressed in Construction \[con:smallperserver\] below. Schemes with small per-server storage {#subsec:perserver} ------------------------------------- We make the observation that the last construction may be used to give families of schemes with lower per-server storage; see [@SRR14 Section V] for similar techniques. The point here is that we never XOR the first bit (say) from one block with the second bit (say) of any other block, so we can store these bits in separate servers without causing problems. More precisely, let $s$ be a fixed integer such that $s\mid R$ and let $t$ be a fixed integer such that $(t-1)\mid s$. We divide each record $X_i$ into $R/s$ blocks $\pi_1(X_i), \pi_2(X_i),\ldots,\pi_{R/s}(X_i)$, each $s$ bits long. We then divide each block $\pi_j(X_i)$ into $(t-1)$ sub-blocks $\pi_{j,1}(X_i)$, $\pi_{j,2}(X_i),\ldots ,\pi_{j,t-1}(X_i)$, each $s/(t-1)$ bits long. For any $i\in\{1,2,\ldots,k\}$ and any $j\in\{1,2,\ldots,R/s\}$, we define $\pi_{j,0}(X_i)$ to be the all zero string $0^{s/(t-1)}$ of length $s/(t-1)$. \[con:smallperserver\] Let $s$ be a fixed integer such that $s\mid R$. Let $t$ be a fixed integer such that $(t-1)\mid s$. Let $n=t(R/s)$. Suppose there are $n$ servers. Each server will store just $ks$ bits. - Index the $t(R/s)$ servers by pairs $(u,r)$, where $1\leq r\leq t$ and where $1\leq u\leq R/s$. Server $S_{(u,r)}$ stores the $u$th sub-block of every block. So $S_{(u,r)}$ stores $\pi_{u,j}(X_i)$ where $1\leq i\leq k$ and $1\leq j\leq t-1$. Note that each server stores $k(t-1)s/(t-1)=ks$ bits. - A user who requires Record $\ell$ chooses $k$ elements $a_1,a_2,\ldots ,a_k\in\mathbb{Z}_t$ uniformly and independently at random. The server $S_{(k,r)}$ is sent the query $q_r=(b_{1r},b_{2r},\ldots,b_{kr})\in\mathbb{Z}_{t}^k$, where $$b_{ir}=\begin{cases} a_i+r\bmod n&\text{if }i=\ell,\\ a_i&\text{otherwise}. \end{cases}$$ (Note that many servers receive the same query.) - Server $S_{(u,r)}$ returns the $s/(t-1)$-bit string $c_{(u,r)}=\oplus_{i=1}^k \pi_{u,b_{ir}}(X_i)$. - To recover the $j$th sub-block of the $u$th block of $X_\ell$, the user finds integers $r$ and $r'$ such that $b_{\ell r}=0$ and $b_{\ell r'}=j$ and computes $c_{(u,r)}\oplus c_{(u,r')}$. \[thm:construction4\] Construction \[con:smallperserver\] is a PIR scheme with download complexity $\frac{R}{s}\frac{r}{t-1}s=\frac{t}{t-1}R$. The scheme has upload complexity $nk\log t=(tkR/s)\log t$ and total storage $nks=tkR$ bits. As in the proofs of Theorems \[thm:construction2\] and \[thm:construction3\], privacy follows since $S_{u,r}$ always receives a uniformly distributed vector $q_r\in\mathbb{Z}_t^k$ as a query. For correctness, observe that when $i\not=\ell$ $$\pi_{u,b_{ir}}(X_i)\oplus \pi_{u,b_{ir'}}(X_i)=\pi_{u,a_{i}}(X_i)\oplus \pi_{u,a_{i}}(X_i)=0^{s/(t-1)},$$ but when $i=\ell$ $$\pi_{u,b_{ir}}(X_i)\oplus \pi_{u,b_{ir'}}(X_i)=\pi_{u,0}(X_i)\oplus \pi_{u,j}(X_i)=\pi_{u,j}(X_i)= \pi_{u,j}(X_\ell).$$ Hence $$c_{(u,r)}\oplus c_{(u,r')}=\bigoplus_{i=1}^k(\pi_{u,b_{ir}}(X_i)\oplus \pi_{u,b_{ir'}}(X_{i}))=\pi_{u,j}(X_{\ell}).$$ So the user can indeed compute the $j$-th sub-block of the $u$-th block as claimed. It is easy to calculate the upload complexity, download complexity and total storage complexity as before, remembering that each server stores $ks$ bits rather than the entire database. By fixing $t$ and $s$ to be sufficiently large integers, we can see that for all positive $\epsilon$ we have a family of schemes with download complexity at most $(1+\epsilon)R$, with total storage linear in the database size, with a linear (in $R$) number of servers, and where the per server storage is independent of $R$. So this family of schemes has a better download complexity and per-server storage than Shah et al. [@SRR14 Section V], and is comparable in terms of both the number of servers and total storage. The servers may be divided into $t$ classes $\mathcal{S}_1,\mathcal{S}_2,\ldots,\mathcal{S}_{t}$, where $$\mathcal{S}_r=\{S_{(1,r)},S_{(2,r)},\ldots,S_{(R/s,r)}\}.$$ Since servers in the same class receive the same query, the above construction still works if some of the servers within a class are merged. If this is done, the storage requirements of each merged server is increased, the download complexity and total storage are unaffected, and the number of servers required and upload complexity are reduced. So various trade-offs are possible using this technique. An explicit asymptotically optimal scheme {#subsec:sunjafar} ----------------------------------------- Sun and Jafar [@SuJa16d] describe a PIR scheme that has the best possible asymptotic download complexity, as $R\rightarrow\infty$. Their scheme is constructed in a recursive fashion. In this subsection, we describe an explicit, non-recursive, scheme with the same parameters as the Sun and Jafar scheme. Our scheme has the advantages of a more compact description, and (we believe) a proof that is significantly more transparent. Our scheme is described in detail in Construction \[con:sunjafar\] below. But, to aid understanding, we first provide an overview of the scheme. Suppose that $n^k$ divides $R$. We split an $R$-bit string $X$ into $n^k$ blocks, each of length $R/n^k$. For $j\in\{1,2,\ldots,n^k\}$ we write $\pi_j(X)$ for the $j$-th block of $X$, and we write $\pi_0(X)$ for the all zero block $0^{R/n^k}$. Let $\mathcal{V}$ be the set of all non-zero strings ${\mathbf{v}}=v_1v_2v_3\ldots v_k\in\{0,1,2,\ldots,n-1\}^k$ such that $\sum_{i=1}^k v_i\equiv 0\bmod n-1$. (Note that our sum is taken modulo $n-1$, not modulo $n$.) Let $\mathcal{W}=\{1,2,\ldots,n\}\times \mathcal{V}$. There are $n$ servers in the scheme, each storing the whole database. Server $S_r$ receives a query consisting of integers $b_i(r,{\mathbf{v}})\in\{1,2,\ldots,n^k\}$ where $i\in\{1,2,\ldots,k\}$ and ${\mathbf{v}}\in \mathcal{V}$. The server replies with $|\mathcal{V}|$ strings, each of length $R/n^k$. Each string is a linear combination of blocks, at most one block from each record (the choice of each block being determined by an integer $b_i(r,{\mathbf{v}})$: see  below). From the perspective of $S_r$, the distribution of the integers $b_i(r,{\mathbf{v}})$ does not depend on $\ell$, enabling us to attain privacy. However, the user chooses these integers so that $b_i(r,{\mathbf{v}})$ and $b_{i}(r',{\mathbf{v}}')$ are sometimes constrained to be equal when $r\not=r'$. This is done in such a way that the user can reconstruct Record $\ell$ from the replies of the servers. To describe the constraints the user imposes, which of course depend on $\ell$, we define a graph $\Gamma^{[\ell]}$ on the set $\mathcal{W}$, and constrain $b_i(r,{\mathbf{v}})$ and $b_{i}(r',{\mathbf{v}}')$ to be equal when $(r,{\mathbf{v}})$ and $(r',{\mathbf{v}}')$ lie in the same component of $\Gamma^{[\ell]}$. (0,0) node\[left\][$(1,101)$]{}– (10,0.5); (0,1) node\[left\][$(3,202)$]{} – (10,0.5) node\[right\][$(2,002)$]{}; (0,2) node\[left\][$(1,110)$]{} – (10,2.5); (0,3) node\[left\][$(3,220)$]{} – (10,2.5) node\[right\][$(2,020)$]{}; (0,4) node\[left\][$(1,112)$]{} – (10,4.5); (0,5) node\[left\][$(3,222)$]{} – (10,4.5) node\[right\][$(2,022)$]{}; (0,6) node\[left\][$(1,211)$]{} – (10,6.5); (0,7) node\[left\][$(2,121)$]{} – (10,6.5) node\[right\][$(3,011)$]{}; (0,8) node\[left\][$(2,211)$]{} – (10,8.5); (0,9) node\[left\][$(3,121)$]{} – (10,8.5) node\[right\][$(1,011)$]{}; (-1,5) ellipse (3 and 8); (11,5) ellipse (3 and 6.5); (0,0) circle (0.08); (0,2) circle (0.08); (0,4) circle (0.08); (0,6) circle (0.08); (0,8) circle (0.08); (0,1) circle (0.08); (0,3) circle (0.08); (0,5) circle (0.08); (0,7) circle (0.08); (0,9) circle (0.08); (10,0.5) circle (0.08); (10,2.5) circle (0.08); (10,4.5) circle (0.08); (10,6.5) circle (0.08); (10,8.5) circle (0.08); (0,10) node\[left\][$(1,200)$]{} circle (0.08); (-1,-0.5) node[$\vdots$]{}; (11,0) node[$\vdots$]{}; (-1,14) node[$\mathcal{W}_1^{[\ell]}$]{}; (11,12.5) node[$\mathcal{W}_2^{[\ell]}$]{}; We now give details of the scheme. We begin by describing the graph $\Gamma^{[\ell]}$ (see Figure \[fig:sjgraph\]) and by detailing some of its structure. Let $\ell\in\{1,2,\ldots,k\}$. Let $\mathcal{W}_1^{[\ell]}$ consist of those elements $(r,{\mathbf{v}})\in \mathcal{W}$ such that $v_\ell\not=0$, and let $\mathcal{W}_2^{[\ell]}$ consist of the elements such that $v_\ell=0$. The graph $\Gamma^{[\ell]}$ is defined on the vertex set $\mathcal{W}$, and is bipartite with parts $\mathcal{W}_1^{[\ell]}$ and $\mathcal{W}_2^{[\ell]}$. We draw at most one edge from each element $(r,{\mathbf{v}})\in \mathcal{W}_1^{[\ell]}$ into $\mathcal{W}_2^{[\ell]}$ as follows. If $v_\ell$ is the only non-zero component of ${\mathbf{v}}$, we draw no edge from $(r,v_\ell)$, so we have an isolated vertex. Suppose two or more components of ${\mathbf{v}}$ are non-zero. We define $\ell_2\in\{1,2,\ldots,k\}$ to be the next component in ${\mathbf{v}}$ after the $\ell$th that is non-zero, taken cyclically. Let $w\in\{1,2,\ldots,n-1\}$ be such that $w\equiv v_\ell+v_{\ell_2}\bmod n-1$. Define ${\mathbf{v}}'=v'_1v'_2\cdots v'_k$ by $$v'_i=\begin{cases} v_i&\text{if }i\in\{1,2,\ldots,k\}\setminus\{\ell,\ell_2\},\\ 0&\text{if }i=\ell,\\ w&\text{if }i=\ell_2. \end{cases}$$ Let $r'\in\{1,2,\ldots,n\}$ be such that $r'\equiv r+v_\ell\bmod n$. We join $(r,{\mathbf{v}})$ to $(r',{\mathbf{v}}')$. Let $\mathcal{C}^{[\ell]}$ be the set of components of $\Gamma^{[\ell]}$. We note that $\Gamma^{[\ell]}$ has exactly $n$ isolated vertices, namely the vectors of the form $(r,{\mathbf{v}})$ where $r\in\{1,2,\ldots,n\}$ and where ${\mathbf{v}}$ is the single vector defined by $$v_i=\begin{cases} 0&\text{if }i\not=\ell,\\ n-1&\text{if }i=\ell. \end{cases}$$ The remaining components in $\mathcal{C}^{[\ell]}$ are stars consisting of a central vertex in $\mathcal{W}^{[\ell]}_2$ and $n-1$ other vertices all lying in $\mathcal{W}^{[\ell]}_1$. Moreover, we note that if $(r,{\mathbf{v}})$ and $(r',{\mathbf{v}}')$ are distinct vertices in the same component of $\Gamma^{[\ell]}$ then $r\not=r'$. We claim that the number of vertices $(r,{\mathbf{v}})\in\mathcal{W}^{[\ell]}_1$ is $n^k$. To see this, we note that there are $n$ choices for $r$, and then $n^{k-1}$ choices for $v_1,v_2,\ldots,v_{\ell-1},v_{\ell+1},\ldots,v_k$. Once these choices are made $v_\ell\in\{0,1,\ldots,n-1\}$ is determined, since $v_\ell\not=0$ and $\sum_{i=1}^kv_i\equiv 0\bmod n-1$. This establishes our claim. Since every component of $\Gamma^{[\ell]}$ contains a vertex in $\mathcal{W}^{[\ell]}_1$, we see that $|\mathcal{C}^{[\ell]}|\leq |\mathcal{W}^{[\ell]}_1|=n^k$. Indeed, the number of components of $\Gamma^{[\ell]}$ is: $$|\mathcal{C}^{[\ell]}|=n+(|\mathcal{W}^{[\ell]}_1|-n)/(n-1)=n(1+(n^{k-1}-1)/(n-1)).$$ \[con:sunjafar\] Suppose that $n^k\mid R$. Suppose there are $n$ servers, each storing the whole database. - A user who requires Record $\ell$ proceeds as as follows. In the notation defined above, for each $i\in\{1,2,\ldots,k\}\setminus\{\ell\}$ the user chooses (uniformly and independently) a random injection $f_i:\mathcal{C}^{[\ell]} \rightarrow \{1,2,\ldots,n^{k}\}$. The user chooses (again uniformly and independently) a random bijection $\psi: \mathcal{W}^{[\ell]}_1\rightarrow \{1,2,\ldots,n^{k}\}$. Define integers $b_i(r,{\mathbf{v}})\in\{0,1,\ldots ,n^k\}$ for $(r,{\mathbf{v}})\in\mathcal{W}$ and $i\in\{1,2,\ldots,k\}$ as follows. If $i\not=\ell$, define $$b_i(r,{\mathbf{v}})=\begin{cases} 0&\text{if }v_i=0,\\ f_i(C)&\text{if }v_i\not=0 \text{ and $(r,{\mathbf{v}})$ lies in the component $C\in\mathcal{C}$}. \end{cases}$$ Note that when $i=\ell$ we have that $v_i\not=0$ if and only if $(r,{\mathbf{v}})\in\mathcal{W}_1^{[\ell]}$. So when $i=\ell$ we may define $$b_i(r,{\mathbf{v}})=\begin{cases} 0&\text{if }v_i=0,\\ \psi((r,{\mathbf{v}}))&\text{if }v_i\not=0. \end{cases}$$ For $r=1,2,\ldots,n$, server $S_r$ is sent the vector $q_r=(b_i(r,{\mathbf{v}}):{\mathbf{v}}\in\mathcal{V}, i\in\{1,2,\ldots,k\})$. - The server $S_r$ replies with the blocks $$\label{eqn:sj_combination} s_{(r,{\mathbf{v}})}=\sum_{i=1}^{k}\pi_{b_i(r,{\mathbf{v}})}(X_i)$$ for all ${\mathbf{v}}\in\mathcal{V}$. - To recover block $j$ of $X_\ell$, the user finds $(r,{\mathbf{v}})=\psi^{-1}(j)\in\mathcal{W}_1^{[\ell]}$. Let $C\in\mathcal{C}^{[\ell]}$ be the component containing $(r,{\mathbf{v}})$. If $|C|>1$, let $(r',{\mathbf{v}}')\in C\cap \mathcal{W}_2^{[\ell]}$. Then (see below for justification) $$\pi_j(X_\ell)=\begin{cases} s_{(r,{\mathbf{v}})}&\text{ if }|C|=1,\text{ and}\\ s_{(r,{\mathbf{v}})}\oplus s_{(r',{\mathbf{v}}')}&\text{ if }|C|>1. \end{cases}$$ \[thm:construction5\] Construction \[con:sunjafar\] is an $n$-server PIR scheme with download complexity $(1-1/n^k)(n/(n-1))R$. The total storage of the scheme is $nkR$. The upload complexity of the scheme is $k^2n^{k}\log n$ bits. We begin by establishing correctness of the scheme. Let $(r,{\mathbf{v}})=\psi^{-1}(j)$ and let $C\in \mathcal{C}^{[\ell]}$ be the component containing $(r,{\mathbf{v}})$. When $|C|=1$ we have $v_i\not=0$ if and only if $i=\ell$ and so $$s_{(r,{\mathbf{v}})}=\sum_{i=1}^{k}\pi_{b_i(r,{\mathbf{v}})}(X_i)=\pi_{b_\ell(r,{\mathbf{v}})}(X_\ell)=\pi_j(X_\ell),$$ the last equality following since $b_\ell(r,{\mathbf{v}})=j$. Hence the user recovers the $j$th block $\pi_j(X_\ell)$ of $X_\ell$ correctly in this case. Suppose now that $C$ contains two or more vertices, so there exists $(r',{\mathbf{v}}')\in C\cap \mathcal{W}_2^{[\ell]}$. When $i\not=\ell$, the values of $b_i(r,{\mathbf{v}})$ and $b_i(r',{\mathbf{v}}')$ are equal, since $(r,{\mathbf{v}})$ and $(r',{\mathbf{v}}')$ lie in the same component $C$ of $\Gamma^{[\ell]}$ and since $v_i=0$ if and only if $v'_i=0$. Moreover, $v_\ell\not=0$ and $v'_\ell=0$. Hence $$\begin{aligned} s_{(r,{\mathbf{v}})}\oplus s_{(r',{\mathbf{v}}')}&=\sum_{i=1}^k \left(\pi_{b_i(r,{\mathbf{v}})}(X_i)\oplus\pi_{b_i(r',{\mathbf{v}}')}(X_i)\right)\\ &=\pi_{b_\ell(r,{\mathbf{v}})}(X_\ell)\oplus \pi_{b_\ell(r',{\mathbf{v}}')}(X_\ell))\\ &=\pi_{\psi((r,{\mathbf{v}}))}(X_\ell)\oplus\pi_{0}(X_\ell))\\ &=\pi_{j}(X_\ell).\end{aligned}$$ So the user recovers the $j$th block $\pi_j(X_\ell)$ of $X_\ell$ correctly in this case also. We have established correctness. We now aim to establish the security of the scheme. Let $\mathcal{A}$ be the set of integer vectors $(a_i({\mathbf{v}})\in\{0,1,\ldots,n^k\}:i\in\{1,2,\ldots,k\}, {\mathbf{v}}\in\mathcal{V})$ with the restrictions that $a_i({\mathbf{v}})=0$ if and only if $v_i=0$, and that for any fixed $i\in\{1,2,\ldots,k\}$ the integers $a_i({\mathbf{v}})$ with $v_i\not=0$ are distinct. Let $r\in \{1,2,\ldots,n\}$ be fixed. The query $q_r=(b_i(r,{\mathbf{v}}):{\mathbf{v}}\in \mathcal{V},i\in\{1,2,\ldots,k\})$ lies in $\mathcal{A}$, since the functions $f_i$ and $\psi$ are injective and since (whether or not $i=\ell$) we have $b_i(r,{\mathbf{v}})=0$ if and only if $v_i=0$. Indeed, the query is uniformly distributed in $\mathcal{A}$. To see this, first note that the functions $f_i$ (for $i\not=\ell$) and $\psi$ are chosen independently. The values $b_\ell(r,{\mathbf{v}})$ for $v_\ell\not=0$ are uniform subject to being distinct since $\psi$ is a randomly chosen bijection. For $i\not=\ell$, the values $b_i(r,{\mathbf{v}})$ for $v_\ell\not=0$ are uniform subject to being distinct, since $f_i$ is a uniformly chosen injection from $\mathcal{C}^{[\ell]}$, and since at most one vertex in any component $C\in\mathcal{C}^{[\ell]}$ has its first entry equal to $r$. Hence the distribution of query $q_r$ is uniform on $\mathcal{A}$ as claimed. Since this distribution does not depend on $\ell$, privacy follows. Each server replies with $|\mathcal{V}|$ strings, each string of length $R/n^k$. Since there are $n$ servers, the download complexity is $nR|\mathcal{V}|/n^k$. So it remains to determine $|\mathcal{V}|$. For $0\leq s\leq k-1$, there are $n^{k-s-1}$ elements $v_1v_2\cdots v_k\in\mathcal{V}$ that begin with exactly $s$ zeros, since we may choose $v_{s+2},v_{s+3},\ldots,v_{k}\in\{0,1,\ldots,n-1\}$ arbitrarily and then $v_{s+1}$ is determined by the fact it is non-zero and $\sum_{j=1}^kv_j\equiv 0 \bmod n-1$. So $$|\mathcal{V}|=\sum_{s=0}^{k-1}n^{k-s-1}=(n^k-1)/(n-1)$$ and the download complexity is $(1-1/n^k)(n/(n-1))R$, as required. We may argue that the total upload complexity is $k^2n^{k}\log n$ as follows. Consider Server $S_r$. The integers $b_i(r,{\mathbf{v}})$ with $v_i=0$ are zero, and so do not need to be sent. There are exactly $kn^{k-1}$ integers $b_i(r,{\mathbf{v}})\in\{1,2,\ldots,n^k\}$ with $i\in\{1,2,\ldots,k\}$ and ${\mathbf{v}}\in\mathcal{V}$ with $v_i\not=0$. (To see this, note that there are $k$ choices for $i$, and $n$ choices for each component ${\mathbf{v}}$ except the $\ell$th. But then $v_\ell$ is determined by the fact that it is non-zero and $\sum_{j=1}^kv_j\equiv 0\bmod n-1$.) Each integer can be specified using $k\log n$ bits, and so the query $q_r$ is $k^2n^{k-1}\log n$ bits long. Since there are $n$ servers, the total upload complexity is $k^2n^{k}\log n$ bits, as required. An averaging technique {#subsec:averaging} ---------------------- The download complexity of both the PIR scheme due to Sun and Jafar [@SuJa16d] and the scheme in Construction \[con:sunjafar\] above is $(1-1/n^k)(n/(n-1))R$. This is only slightly smaller than the more practical scheme in Construction \[con:smallnumservers\], which has download complexity $(n/(n-1))R$. In fact, the *expected* number of bits downloaded in Construction \[con:smallnumservers\] is $(1-1/n^k)(n/(n-1))R$, since a server is asked for an all-zero linear combination of blocks with probability $1/n^k$ and need not reply in this case. This section describes an ‘averaging’ technique which transforms Construction \[con:smallnumservers\] into a scheme with good (worst case) download complexity, at the price of a much stronger divisibility constraint on the length of blocks. This technique will work for a wide range of PIR schemes, but in the case of Construction \[con:smallnumservers\] it produces a scheme with optimal download complexity $(1-1/n^k)(n/(n-1))R$. Moreover, the upload complexity is considerably smaller than the schemes described in [@SuJa16d] and Construction \[con:sunjafar\]. Before giving the detail, we describe the general idea. Chan, Ho and Yamamoto [@CHY15 Remark 2] observed that a PIR scheme with good upload complexity (but long record lengths) can be constructed by dividing each record into blocks, then using copies of a fixed PIR scheme for shorter records operating on each block in parallel. Crucially, the same randomness (and so the same queries) can be used for each parallel copy of the scheme, and so upload complexity is low. The ‘averaging’ construction operates in a similar way. However, rather than using the same randomness we use different but predictably varying randomness for each parallel copy. The server can calculate queries for each copy of the scheme from just one query, so upload complexity remains low. But (because queries vary over all possibilities) the resulting scheme has (worst case) download complexity equal to the average number of bits of download in the Chan, Ho and Yamamoto construction. In more detail, we modify Construction \[con:smallnumservers\] as follows. Suppose that $n^k(n-1)\mid R$. We divide an $R$-bit string $X$ into $n^k(n-1)$ blocks, each of size $R/(n^k(n-1))$. We index these blocks by pairs $(b,{\mathbf{x}})$ where $b\in\{1,2,\ldots,n-1\}\subseteq{\mathbb{Z}}_n$ and ${\mathbf{x}}\in{\mathbb{Z}}_n^k$. We write $\pi_{(b,{\mathbf{x}})}(X)$ for the block of $X$ that is indexed by $(b,{\mathbf{x}})$. For any ${\mathbf{x}}\in {\mathbb{Z}}_n^k$, we write $\pi_{(0,{\mathbf{x}})}(X)$ for the all-zero string $0^{R/(n^k(n-1))}$ of length $R/(n^k(n-1))$. \[con:averaging\] Let $n$ be an integer such that $n^k(n-1)\mid R$. Suppose there are $n$ servers, each storing the entire database. - A user who requires Record $\ell$ chooses $k$ elements $a_1,a_2,\ldots,a_k\in\mathbb{Z}_{n}$ uniformly and independently at random. For $r=1,\ldots,n$, server $S_r$ is sent the vector $q_r=(b_{1r},b_{2r},\ldots,b_{kr})\in\mathbb{Z}_{n}^k$, where $$b_{ir}=\begin{cases} a_i+r\bmod n&\text{if }i=\ell,\\ a_i&\text{otherwise}. \end{cases}$$ - For $r\in\{1,2,\ldots,n\}$ and ${\mathbf{x}}\in{\mathbb{Z}}_n^k$, define the string $c_{(r,{\mathbf{x}})}$ of length $R/(n^k(n-1))$ by $$c_{(r,{\mathbf{x}})}=\bigoplus_{i=1}^k \pi_{(b_{ir}+x_i,{\mathbf{x}})}(X_i).$$ The server $S_r$ returns the string $c_{(r,{\mathbf{x}})}$, for all ${\mathbf{x}}=(x_1,x_2,\ldots,x_k)\in{\mathbb{Z}}_n^k$ such that ${\mathbf{x}}+q_r\not=\mathbf{0}$. So $S_r$ returns $n^k-1$ strings. - To recover the block of $X_\ell$ indexed by a pair $(j,{\mathbf{x}})$, the user finds the integers $r$ and $r'$ such that $b_{\ell r}+x_\ell=0$ and $b_{\ell r'}+x_{\ell}=j$. The user then computes $c_{(r,{\mathbf{x}})}\oplus c_{(r',{\mathbf{x}})}$. \[thm:construction6\] Construction \[con:averaging\] is an $n$-server PIR scheme with download complexity $(1-1/n^k)\frac{n}{n-1}R$. The scheme has upload complexity $nk\log n$ and total storage is $nkR$. We begin with the correctness of the scheme. Exactly as in the proof of Theorem \[thm:construction3\], we note that $r$ and $r'$ exist since $b_{\ell r}+x_\ell\in\{0,1,2\ldots,n-1\}$ takes on each possible value once as $r\in\{0,1,\ldots,n\}$ varies. Moreover, we note that the string $c_{(r,{\mathbf{x}})}$ is all zero if ${\mathbf{x}}+q_r=0$ (and similarly the string $c_{(r',{\mathbf{x}})}$ is all zero if ${\mathbf{x}}+q_{r'}=0$) and so the user always receives enough information to calculate $c_{(r,{\mathbf{x}})}\oplus c_{(r',{\mathbf{x}})}$. Let ${\mathbf{x}}=(x_1,x_2,\ldots,x_k)$. When $i\not=\ell$ $$\pi_{(b_{ir}+x_i,{\mathbf{x}})}(X_i)\oplus \pi_{(b_{ir'}+x_i,{\mathbf{x}})}(X_i)=\pi_{(a_{i}+x_i,{\mathbf{x}})}(X_i)\oplus \pi_{(a_{i}+x_i,{\mathbf{x}})}(X_i)=0^{R/(n-1)}.$$ When $i=\ell$ $$\pi_{(b_{ir}+x_i,{\mathbf{x}})}(X_i)\oplus \pi_{(b_{ir'}+x_i,{\mathbf{x}})}(X_i)=\pi_{(0,{\mathbf{x}})}(X_i)\oplus \pi_{(j,{\mathbf{x}})}(X_i)=\pi_{(j,{\mathbf{x}})}(X_i)= \pi_{(j,{\mathbf{x}})}(X_\ell).$$ Hence $$c_{(r,{\mathbf{x}})}\oplus c_{(r',{\mathbf{x}})}=\bigoplus_{i=1}^k(\pi_{(b_{ir}+x_i,{\mathbf{x}})}(X_i)\oplus \pi_{(b_{ir'}+x_i,{\mathbf{x}})}(X_{i}))=\pi_{(j,{\mathbf{x}})}(X_{\ell}).$$ So the user recovers the block of $X_{\ell}$ indexed by $(j,{\mathbf{x}})$ correctly. Privacy follows from the privacy of Construction \[con:smallnumservers\], as the method for generating queries is identical. The total storage is $nkR$, since each of $n$ servers stores the entire $kR$-bit database. Each query $q_r$ is $k\lceil \log n\rceil $ bits long, since an element of $\mathbb{Z}_n$ may be specified using $ \log n$ bits. Hence the upload complexity is $nk\log n$. Since there are $n$ servers, and each server returns $n^k-1$ strings of length $R/(n^k(n-1))$, the download complexity is $(1-1/n^k)\frac{n}{n-1}R$. Conclusions and future work {#sec:conclusion} =========================== In this paper, we have used classical PIR techniques to prove bounds on the download complexity of PIR schemes in modern models, and we have presented various constructions for PIR schemes which are either simpler or perform better than previously known schemes. Various interesting problems remain in this area. We first consider schemes with optimal download complexity: Are there PIR schemes with fewer than $R+1$ bits of download complexity? Our paper, like the rest of the literature, only considers PIR schemes over binary channels, and in this model the answer is ‘no’. But the proofs of this fact in this paper and in Shah at el. [@SRR14] both use the fact that we are working over binary channels: more than $R$ bits of download implies that at least $R+1$ bits are downloaded. So this problem is still open if we extend the model to schemes that do not necessarily use binary channels. We now return to the standard binary channel model. \[prob:Shah\] Are there PIR schemes with download complexity $R+1$ and total storage linear in $R$? This result was claimed in Shah at el. [@SRR14], but we believe that a proof of this is still not known. A proof of this result might depend on a more detailed structural analysis of PIR schemes with $R+1$ bits of download. As a first step, we believe the following to be of interest: Theorem \[thm:almostall\] bounds the probability that only $R$ bits are downloaded in a PIR scheme with (worst case) download complexity $R+1$. Is this bound tight? We conjecture that the bound could be significantly improved in some cases. We now consider families of schemes that have good asymptotic complexity as $R\rightarrow\infty$. Does there exist a family of schemes with download complexity $(1+o(R))R$ and linear total storage? Note that an affirmative solution to Question \[prob:Shah\] will imply an affirmative solution to this question. Are there practical PIR schemes that approach asymptotic capacity as $R$ grows? The schemes by Sun and Jafar [@SuJa16b] and the schemes presented in this paper have the strong restriction that $n^k$ must divide $R$. This makes the schemes impractical for many parameters. Is there a combinatorial proof that provides a tight upper bound on the asymptotic capacity as $R\rightarrow\infty$? We comment that the proof in Sun and Jafar [@SuJa16b] uses information theoretic techniques. A combinatorial proof might give extra structural information for schemes meeting the bound, and might improve the bound in non-asymptotic cases. Finally, we turn to larger questions. Can we find better constructions for PIR schemes? Schemes are of interest if they improve per server storage, total storage, upload or download complexity, if the number of servers needed was reduced, or if the divisibility conditions for parameters such as $R$ are weakened. Can the techniques from this paper be applied to establish bounds or give constructions in other models, such as those discussed in Subsection \[subsec:context\]? #### Acknowledgement The authors would like to thank Doug Stinson for comments on an earlier draft. [99]{} , [*Nearly optimal constructions of PIR and batch codes,*]{} [*Proc. of IEEE Int. Symp. on Inform. Theory (ISIT)*]{}, pp.151–155, Aachen, Germany, June 2017. in [*Cryptology and Network Security,*]{} pp.222–239, Springer 2014. , September 2016. , February 2017. , pp.1898–1902, Aachen, Germany, June 2017. , June 2017. , 57 (2011), 1479–1494. , August 2016. , pp.2658–2662, Aachen, Germany, June 2017. , pp.146–150, Aachen, Germany, June 2017. , October 2014. , pp.2842–2846, Hong Kong, June 2015. , 45 (1998), 965–981. , 56 (2010), 4539–4551. , 56 (2011), 476–489. , [*Multi-server private information retrieval over unsynchronized databases,*]{} [*Fifty-second Annual Allerton Conference*]{}, pp.437–444, Illinois, October 2014. , 9 (2015), 1229–1239. , pp.2852–2856, Hong Kong, June 2015. , May 2015. , April 2017. , [*Private information retrieval from coded databases with colluding servers,*]{} [*arxiv.org/abs/1611.02062*]{}, November 2016. , 58 (2012), 6925–6934. , [*Linear locally repairable codes with availability,*]{} [*Proc. of IEEE Int. Symp. on Inform. Theory (ISIT)*]{}, pp.1871–1875, Hong Kong, June 2015. , pp.262–271, Chicago, IL, June 2004. , September 2017. , January 2017. , January 2017. , pp.364–373, Miami Beach, FL, October 1997. , [*Lengthening and extending binary private information retrirval codes,*]{} [*arxiv.org/abs/arXiv:1707.03495*]{}, July 2017. , [*Lower bound on the redundancy of PIR codes,*]{} [*arxiv.org/abs/1605.01869*]{}, May 2016. , pp.631–635, Honolulu, HI, June 2014. , 62 (2016), 4481–4493. , 62 (2016), 1592–1604. , January 2017. , pp.829–833, Aachen, Germany, June 2017. , [*One extra bit of download ensures perfectly private information retrieval,*]{} in [*Proc. of IEEE Int. Symp. on Inform. Theory (ISIT)*]{}, pp.856–860, Honolulu, June 2014. , 63 (2017), 4075–4088. , May 2016. , June 2016. , October 2016. , November 2016. , January 2017. , pp.1893–1897, Aachen, Germany, June 2017. , February 2016. , pp.1411–1415, Barcelona, Spain, July 2016. , pp.1903–1907, Aachen, Germany, June 2017. , January 2017. , [*The capacity of cache aided private information retrieval,*]{} [*arxiv.org/abs/arXiv:1706.07035*]{}, June 2017. , [*Binary, shortened projective Reed Muller codes for coded private information retrieval,*]{} [*arxiv.org/abs/arXiv:1702.05074*]{}, February 2017. , [*Binary, shortened projective Reed Muller codes for coded private information retrieval,*]{} [*Proc. of IEEE Int. Symp. on Information Theory (ISIT)*]{}, pp.2653–2657, Aachen, Germany, June 2017. , October 2016. , July 2017. , August 2017. , 53 (2010), 68–73. , [*A general private information retrieval from MDS coded databases with colluding servers,*]{} [*arxiv.org/abs/1704.06875*]{}, April 2017. , [*Multi-file private information retrieval scheme for MDS coded databases with colluding servers,*]{} [*arxiv.org/abs/1704.06875*]{}, May 2017. , [*On private information retrieval array codes,*]{} [*arxiv.org/abs/1609.09167*]{}, September 2016. , [*Private multi-file retrieval from distributed databases,*]{} [*arxiv.org/abs/1704.00250*]{}, April 2017. [^1]: Department of Mathematics, Royal Holloway University of London, Egham, Surrey TW20 0EX, United Kingdom, e-mail: [s.blackburn@rhul.ac.uk]{}. [^2]: Department of Computer Science, Technion, Haifa 3200003, Israel, e-mail: [etzion@cs.technion.ac.il]{}. Part of the research was performed while the second author visited Royal Holloway University of London under EPSRC Grant EP/N022114/1. Part of the research was supported by the NSF-BSF grant 2016692. [^3]: Department of Economics, Mathematics and Statistics, Birkbeck, University of London, Malet Street, London WC1E 7HX, United Kingdom, e-mail: `m.paterson@bbk.ac.uk`. [^4]: Parts of this paper were presented at the International Symposium on Information Theory, Aachen, Germany, June 2017

Ret mouse very large tumors (VLTs) display altered ratios of infiltrating memory to naive T cells: Roles in tumor expansion. Melanoma is an aggressive skin cancer, however it is immunogenic. The size of the primary tumor is associated with the nodal metastases. Our goals were to characterize melanoma-associated antigens (MAAs) and tumor-infiltrating T-lymphocytes (TILs) subsets in the few very large tumors (VLTs) developing in ret transgenic mice of melanoma. Tumors >700mg (VLTs) were investigated for MAAs and subsets of TILs. Immunohistochemistry and flow cytometry-based studies were performed to determine the infiltration patterns of T-lymphocytes in VLTs. It was observed that zinc fixative restores the antigenicity of the cell-surface markers of lymphocyte subpopulations without the need of antigen retrieval, whereas formalin-based fixative fails to restore the antigenicity in the presence of antigen retrieval in the immunohistochemistry. VLTs from ret mice express MAAs, such as Tyrosinase, TRP-1, TRP-2 and gp-100. The mean±standard deviation (S.D.) T-cell infiltration per 400 times-high power field in VLTs; CD4(+) (2.33±1.3), CD8(+) (2.00±1.0), and CD4(+) Foxp3(+) (2.5±0.5) regulatory T cells infiltration was exclusively restricted to the tumor stroma. Moreover, our flow cytometry-based data reveal that % mean±S.D. naive CD3(+) CD4(+) T cell infiltration (32.8±4.0%) was significantly larger than effector (25.8±2.8%, p<0.01) and central memory cells (16.1±3.7%, p<0.001) in VLTs. Similarly, between CD3(+) CD8(+) T cells, naive cells infiltrate (57.7±2.3%) in a significantly larger frequency than effector (5.0±0.4%, p<0.0001) and central memory cell (4.8±1.7%, p<0.0001) subsets. These results suggest that the VLTs from ret mice display lowered infiltration ratios between memory and naive T cells, which could be associated with the relatively large growth of VLTs.

The proposal submitted is a Clinical Cancer Education Program that will be administered by the Office of Education. The Clinical Programs will be in the following areas and will enable us to: 1. Conduct graduate education programs in: a. Clinical Associates Program; b. Oncology Nursing for Clinical Nursing Specialists; c. Cancer Coordinator Program; d. Visiting Professorships. 2. Conduct undergraduate education programs in: a. Clinical Assistantships for Medical Students; b. Clinical Assistantships for Nursing. 3. Conduct continuing education in: a. Short courses in oncology for practicing physicians and dentists; b. Oncological nursing for registered professional nurses. 4. Conduct inservice education in: a. Clinical nursing practice development program. In all of the above programs, the overall objective will be to provide a sound grasp of fundamental information and basic skills needed to perform good cancer management (detection, diagnosis, treatment, rehabilitation, and continuing care). It is also hoped that a mature attitude toward the cancer patient can be expanded by aiding the cancer patient in self assessment of his disease.

History of salt in Middlewich Middlewich, a town in northwest England, lies on the confluence of three rivers – the Dane, the Croco and the Wheelock. Most importantly for the history of salt making, it also lies on the site of a prehistoric brine spring. Following the Roman invasion, Middlewich was named Salinae on account of the salt deposits around it, as it was one of their major sites of salt production. During this time the Romans built a fort at Harbutts Field (SJ70216696), to the north of the town. Recent excavations to the south of the fort have found evidence of further Roman activity including a well and part of a preserved Roman road. Salt manufacture has remained the principal industry for the past 2,000 years. Salt making is mentioned in the Domesday book, and by the 13th century there were approximately 100 "wich houses" packed around the town's two brine pits. By 1908 there were nine industrial scale salt manufacturers in the town, with a number of open pan salt works close to the canal, however salt manufacture in Middlewich is now concentrated in one manufacturer, British Salt. The salt is sold as the Saxa brand by RHM, and by others e.g. supermarket own brands. Salt produced by British Salt in Middlewich has 57% of the UK market for salt used in cooking. Historical extracts concerning salt production in Middlewich Camden's “Magna Brittania”, 1580 "From thence runneth Wever down by Nantwich, not far from Middlewich, and so to Northwich. These are very famous Salt-Witches, five or six miles distant, where brine or salt water is drawn out of pits, which they pour not upon wood while it burneth as the ancient Gauls and Germans were wont to do, but boil it over a fire to make salt thereof. Neither doubt I that these were known unto the Romans, and that from hence was usually paid the Custom of Salt, called Salarium. "For, there went a notable highway from Middlewich to Northwich, raised with gravel to such a height, that a man may easily acknowledge that it was the work of the Romans, seeing that all this country over, gravel is so scare: and from then at this day it is carried to private men’s uses. Mather Paris writeth that Henry III stopped up these salt-pits when in hostile manner he wasted this shire; because the Welshmen, so tumultuous in those days should not have any victuals or provisions from thence. But when the fair beams of peace began once more to shine out, they were opened again. "Then runneth the Dane under Kinderton, the old seat of the ancient race of the Venebles; who, ever since the first coming of the Normans have been commonly called the Barons of Kinderton. Beneath this southwards, the little river croco, runnerth also into the Dan...Croke, the river aforesaid being past Brereton, within a little while visiteth Middlewich, very near unto his confluence with Dan, where there be two wells of salt water, parted one from the other by a small brook: Sheathes they call them. The one stands not open, but at certain set times, because folk willingly steal the water thereof, as being of greater virtue and efficacy. From thence runneth Dan to Bostoke, in time past Botestock, the ancient sear of the family of the Bostokes, Knights. Out of this ancient house of the Bostoks, as out of a stock, sprung a goodly number of the same name, in Cheshire, Shropshire, Berkshire and elsewhere." Longitudinal study of salt sites in Middlewich Croxton Works The Croxton Works were located on the Trent and Mersey canal, approximately halfway between the Big Lock and the Croxton Lane Bridge at SJ699669. The works were established by the Dairy and Domestic Salt Company, probably in 1892. It was taken over by Henry Seddon before 1905 and worked until closed by subsidence in the 1920s. Until the early 1990s a derelict canal side warehouse still existed on the site, however this has now been demolished. All that remains now is a canal side flash (proposed as the site of a Middlewich Marina in the 1970s) and the foundations for the warehouse. Both the flash and warehouse foundations are now overgrown and hardly visible. Kinderton Salt Works It is likely that this is the only saltworks next to the Roman fort on Harbutt's field. Salt making sites in Cheshire places this site at SJ703668, however the 1882 Ordnance Survey map places the salt pans at approximately SJ7032266605, whilst Middlewich 900–1900 mentions the salt workings being yards away from the stone houses off King Street (i.e. the location given in the 1882 OS map). It is likely that this was the salt works of the Baron of Kinderton, Peter Venables, in 1671, and it is listed in documents of 1682 as producing a weekly output of 2,210 bushels of salt from its seven pans. By the mid-eighteenth century this was the only saltworks on the Kinderton side of the River Croco. In the mid nineteenth century Ralph Seddon owned the works, and on his death it was sold to the Salt Union in 1888. Sometime between 1888 and 1919 the site was dismantled, however a capped off shaft which once formed part of the works could be seen from the path running from King Street to the Big Lock until the new housing estate was built. Pepper Street Salt Works In around 1913 the Pepper Street salt works were rebuilt by Henry Seddon. Following a merger between Seddon and Sons and Cerebos in the late 1950s, the open pans at Pepper Street and those at the Cerebos site on Booth Lane were worked together as a single department, before being closed in 1968–1970. The Pepper Street works were demolished in the mid 1970s (at around the same time as the gas works on the opposite side of the canal), and the site is now a housing estate. Wych House Lane Salt Works In the 14th century the area around the current Wych House Lane was occupied by many salt houses. In 1892 a new salt works was established to the north of Wych House Lane, owned by the Dairy and Domestic Company. In common with many of the works this was taken over by Henry Seddon in the early 1900s. The works continued to be used until around 1969, and were used by the town council as a depot until the 1980s. The land is currently a green field running down to the canal. Chesworth’s/Newtons Salt Works Salt making sites in Cheshire locates Newtons Salt Works at the same site as Wych House Lane Salt works (SJ705662). However the 1898 map places Newtons Salt Works to the south side of Wych House Lane, at approximately SJ706661. Aman's Salt Works Aman's Salt Works was opened on Brooks Lane shortly after the discovery in 1889 of rock salt and brine at the adjacent Murgatroyd's site, with the earliest entry in the accounts book being the 16 November 1892. The location of Aman's Works, between the Trent and Mersey canal and the railway branch line between Sandbach and Northwich, is indicative of the move from canals to railways for transport during the nineteenth century. To this end the works had its own siding and platform for loading trains adjacent to the branch line (around 300 metres from Middlewich railway station). Murgatroyd's Salt works Brunner Mond Salt Works – Brooks Lane British Salt The only remaining salt works in Middlewich is the British Salt works at Cledford (SJ716644). This salt works obtains its brine from Warmingham nearby, rather than Middlewich. Salt from this works is sold by RHM under the Saxa brand. Regeneration All of the old town centre salt works are now closed. Because of the aggressive chemicals that were handled, salt houses were 'temporary' structures, and so unlike the mills constructed in the mill towns of Lancashire, they would be unsuitable for conversion into other uses. Consequently, almost all the structures of the old salt works have been pulled down and the land has been put to other uses. See also History of salt Salt in Cheshire Open pan salt making Weaver Hall Museum and Workhouse, Northwich References External links Salt making sites in Cheshire, George Twigg Salt Manufacturers Association: history of salt History of the Murgatroyd's salt works Middlewich Middlewich Salt

Shawnee, Kansas Welcomes Mobile Window Tint Service The mobile window tint in Shawnee, Kansas is gaining public attention. People are willing to shell out their money in exchange of unique customer service experiences. They demand to know how they will benefit if they avail the mobile service. They also search for all options and select the best. The window tinting near me in Shawnee knows how to spell the best. Customers benefit in a mobile window tint in Shawnee because it provides an alternative, a high rating performance, and a unique experience. It also offers a deserving fee, and most especially, it is conscious of window tinting guidelines. Breaking Barriers to Offer Alternative Mobile Service Every customer deserves to have the best option. If people bring their cars to shops to get the service they need, a breakaway from old school is taking place. The mobile tint service near me in Shawnee, Kansas, is offering an alternative. This kind of service is breaking barriers to offer an alternative way getting your car window tinted. Excellent Service Performance Is Being Sought the Most People do not settle for an ‘okay’ service anymore. They seek for something excellent or perfect. They thirst for what they think they deserve. At mobile window tint in Shawnee, Kansas, premium service is not only delivered. Personalized service is given with sincerity and consistency. This Is Not About You,it Is About Me That is correct! This is not about the mobile car tint service in Shawnee, Kansas. This is not about ensuring that every car gets tinted for some reasons. This is not about squeezing profits from customers. It is all about “the customers” because they are the boss. They define what they need and want. Laws Are There for a Reason, They’re not Meant to be Broken Laws that are enacted and implemented affect people’s lives. Because laws have purposes, people recognize and adhere to them. The window tinting laws of Shawnee, Kansas, are being observed even in mobile service. No customers would just hire and have their cars tinted. High standards mean compliance with the laws. Quality Service Matters but Cost Is Part of the Equation People nowadays are practical. They know that they should spend wisely and smartly. Customers also search for quality service. At mobile car window tinting near me in Shawnee, Kansas, both quality and affordability are part of the equation. The two are inseparable for a positive customer service experience. There you go! There is no more reason to doubt or be worried. Everything is taken care of, because at mobile window tint in Shawnee, Kansas, the equation is fully solved.

Marc Krauss Krauss opened 2014 as a regular in the Astros' lineup, drawing starts between first base and the corner outfield positions while the team waited on Jonathan Singelton in the minors. The former second-... Krauss, a left-handed hitting outfielder, made his major league debut with the Astros in June after displaying good power potential in the minors. The 26-year-old accumulated 146 plate appearances, but only hit four home runs and struck out 30.8% of the time after showing solid plate discipline at the Triple-A level. Krauss will likely get another opportunity in Houston this season as a left-handed bat off the bench or DH, but it remains to be seen whether he will break camp with the club or if it will take an injury to a starter to get that opportunity. 2011 Krauss had little difficulty mashing in the High-A California League last season, swatting 25 homers and ranking sixth in the league in that category. Defensively, he's better suited to play left field than right, but Krauss was given some time there in the Arizona Fall League and didn't exactly draw rave reviews. That said, the combination of raw power, on-base skills and ability to hit to the opposite field are enough to make him intriguing. He'll open 2011 at Double-A Mobile, where more advanced pitching and fewer hitter-friendly environments should paint a clearer picture about his his big league future. If his defense is deemed adequate, there may not be much standing in his way for the left field gig in Arizona with the organization souring on Gerardo Parra as an everyday player.

What to read next? Life in Ancient Rome Award-winning historian Lionel Casson paints a vivid portrait of life in ancient Rome – for slaves and emperors, soldiers and commanders alike – during the empire’s greatest period, the first and second centuries A.D.

Structure of vibriophage D10 tail sheath revealed by electron microscopy and computational image processing. Vibriophage D10, a member of the Vibrio cholerae O-1El-Tor phage typing scheme, is used to detect the spread of the cholera epidemic and belongs to the Myoviridae family. The outer sheath of the tail of vibriophage is highly contractile in nature. We have used electron microscopy and computational image-processing techniques to determine the structure of this contractile tail sheath. The three-dimensional density map of the tail sheath reveals the presence of ∼35 Å long and ∼25 Å wide protrusions, extending out of the tail structure. The electron micrographs revealed that the tail sheaths of a considerable number of D10 phage particles undergo axial compression up to 51% at almost neutral pH (7.2) and at room temperature (20°). We find that the genome of the phage particles is ejected out when the tail sheath of the phage particles are deliberately made to contract by subjecting them to a surrounding environment of pH 10.5. We infer that the contraction of the tail sheath is responsible for the loss of the phage genome even at neutral pH and room temperature. This may be a plausible reason for the unusual behavior of rapid decline of the phage within a span of 48-72 h of its preparation.

A Death in the Desert "A Death in the Desert" is a short story by Willa Cather. It was first published in The Scribner's in January 1903. Plot summary Everett is on a train from Holdrege, Nebraska to Cheyenne, Wyoming. He is a man that looks like his older prodigy brother Adriance—this similarity haunts him throughout the entire novel and robs him of his own personality. He will always be Adriance's brother. Characters Everett Hilgarde Charley Gaylord Two girls on the train. Adriance Hilgarde Katharine Gaylord Maggie Gaylord The Parson Diana, a chaste actress in New York City. Allusions to other works Katharine Gaylord mentions Richard Wagner's Das Rheingold, Edward Gibbon's The History of the Decline and Fall of the Roman Empire, and Heinrich Heine's Florentine Nights. Moreover, Gussie Davis's song In The Baggage Coach Ahead is mentioned - albeit 'in' is elided. Literary significance and criticism It has been argued that the title of the story was influenced by Willa Cather's reading of Robert Browning. Allusions to Alexandre Dumas, fils' La dame aux camelias and Lucretius's De rerum natura have also been found. References External links Full Text at the Willa Cather Archive Category:1903 short stories Category:Short stories by Willa Cather Category:Works originally published in Scribner's Magazine

Monosaccharide nomenclature Monosaccharide nomenclature is a set of conventions used in chemistry to name the compounds known as monosaccharides or "simple sugars" — the basic structural units of carbohydrates, which cannot be hydrolysed into simpler units. Systematic name of molecular graph The elementary formula of a simple monosaccharide is CnH2nOn, where the integer n is at least 3 and rarely greater than 7. Simple monosaccharides may be named generically according on the number of carbon atoms n: trioses, tetroses, pentoses, hexoses, etc. Every simple monosaccharide has an acyclic (open chain) form, which can be written as H-(CH(OH))_\mathit{x}-(C=O)-(CH(OH))_\mathit{y}-H; that is, a straight chain of carbon atoms, one of which is a carbonyl group, all the others bearing a hydrogen -H and a hydroxyl -OH each, with one extra hydrogen at either end. The carbons of the chain are conventionally numbered from 1 to n, starting from the end which is closest to the carbonyl. If the carbonyl is at the very beginning of the chain (carbon 1), the monosaccharide is said to be an aldose, otherwise it is a ketose. These names can be combined with the chain length prefix, as in aldohexose or ketopentose. Most ketoses found in nature have the carbonyl in position 2; when that is not the case, one uses a numeric prefix to indicate the carbonyl's position. Thus for example, aldohexose means H(C=O)(CHOH)5H, ketopentose means H(CHOH)(C=O)(CHOH)3H, and 3-ketopentose means H(CHOH)2(C=O)(CHOH)2H. An alternative nomenclature uses the suffix '-ose' only for aldoses, and '-ulose' for ketoses. The position of the carbonyl (when it is not 1 or 2) is indicated by a numerical infix. For example, hexose in this nomenclature means H(C=O)(CHOH)5H, pentulose means H(CHOH)(C=O)(CHOH)3H, and hexa-3-ulose means H(CHOH)2(C=O)(CHOH)3H. Naming of acyclic stereoisomers Open-chain monosaccharides with same molecular graph may exist as two or more stereoisomers. The Fischer projection is a systematic way of drawing the skeletal formula of an open-chain monosaccharide so that each stereoisomer is uniquely identified. Two isomers whose molecules are mirror-images of each other are identified by prefixes '-' or '-', according to the handedness of the chiral carbon atom that is farthest from the carbonyl. In the Fischer projection, that is the second carbon from the bottom; the prefix is '-' or '-' according to whether the hydroxyl on that carbon lies to the right or left of the backbone, respectively. If the molecular graph is symmetrical (H(CHOH)(CO)(CHOH)H) and the two halves are mirror images of each other, then the molecule is identical to its mirror image, and there is no '-' form. A distinct common name, such as "glucose" or "ribose", is traditionally assigned to each pair of mirror-image stereoisomers, and to each achiral stereoisomer. These names have standard three-letter abbreviations, such as 'Glc' for glucose and 'Rib' for ribose. Another nomenclature uses the systematic name of the molecular graph, a '-' or '-' prefix to indicate the position of the last chiral hydroxyl on the Fischer diagram (as above), and another italic prefix to indicate the positions of the remaining hydroxyls relative to the first one, read from bottom to top in the diagram, skipping the keto group if any. These prefixes are attached to the systematic name of the molecular graph. So for example, -glucose is -gluco-hexose, -ribose is -ribo-pentose, and -psicose is -ribo-hexulose. Note that, in this nomenclature, mirror-image isomers differ only in the ''/'' prefix, even though all their hydroxyls are reversed. The following tables shows the Fischer projections of selected monosaccharides (in open-chain form), with their conventional names. The table shows all aldoses with 3 to 6 carbon atoms, and a few ketoses. For chiral molecules, only the '-' form (with the next-to-last hydroxyl on the right side) is shown; the corresponding forms have mirror-image structures. Some of these monosaccharides are only synthetically prepared in the laboratory and not found in nature. Names of aldoses Names of ketoses Names of 3-ketoses Cyclic forms For monosaccharides in their cyclic form, an infix is placed before the '-ose', '-ulose', or n-ulose' suffix to specify the ring size. The infix is "furan" for a 5-atom ring, "pyran" for 6, "septan" for 7, and so on. Ring closure creates another chiral center at the anomeric carbon (the one with the hemiacetal or acetal functionality), and therefore each open-chain stereoisomer gives rise to two distinct stereoisomers (anomers). These are identified by the prefixes 'α-' and 'β-', which denote the relative configuration of the anomeric carbon to that of the stereocenter at the other end of the carbon chain. To determine if the sugar is α or β, the structure is drawn in a Fischer projection; if the endocyclic oxygen (O5) and exocyclic oxygen (O1) are cis, the sugar is α; if they are trans'', the sugar is β.Trivial Nomenclature of Monosaccharides Examples Glycosides Glycosides are saccharides in which the hydroxyl -OH at the anomeric centre is replaced by an oxygen-bridged group -OR. The carbohydrate part of the molecule is called glycone, the -O- bridge is the glycosisdic oxygen, and the attached group is the aglycone. Glycosides are named by giving the aglyconic alcohol HOR, followed by the saccharide name with the '-e' ending replaced by '-ide'; as in [[phenol D-glucopyranoside]]. Modified Sugars Deoxy sugars Modification of sugar is generally done by replacing one or more –OH group with other functional groups at all positions except C-1. Since all these cases involves the removal of an –OH group, they are all deoxy sugars.Rules for Nomenclature of Modified Sugars: State the Sugar is deoxy sugar. Specify the position of deoxygenation. If there is a substituent other than H in the place of –OH, specify what it is. Specify the relative configuration of all stereogenic centres (manno, gluco etc.). Specify the ring size (furanose, pyranose etc.) and anomeric configuration ( a or b). State the chain length only in situation where –OH is replaced with H. Alphabetize all the substituent groups (deoxy, -iodo, -amino etc.). Di-, tri- etc. prefixes do not count. Examples Protected Sugars Sugars in which –OH is protected by some modification are called protected sugars.Rules for Nomenclature for Protected Sugars:''' Specify the number of particular protecting groups (di, tri, tetra etc.). List groups alphabetically along with all other substituents ( di, tri prefixes do not count). See also Carbohydrate conformation Symbol Nomenclature For Glycans Polysaccharide Oligosaccharide Oligosaccharide nomenclature References Category:Chemical nomenclature Category:Carbohydrates Category:Carbohydrate chemistry