=== asac__ is now known as asac [00:10] persia you here? [00:10] (idle 9+ hours) [00:11] I’d guess he’ll be awake within an hour. :-) [00:12] I know because i remember our sleep rhythms have a phase difference of about 12 hours. :-) [00:13] Uh, that was a brainfart. Going to sleep when someone else wakes up doesn’t mean that. [00:13] crimsun: could you please re-check http://revu.tauware.de/details.py?package=mozilla-devscripts ? I've fixed the FSF address [00:18] Ubulette: looks good. Archived, uploaded. Thanks for your work! [00:18] thx [00:19] will it be in main or universe ? it's a build-dep for some packages in main [00:19] well, s/it's/it will be/ [00:19] obviously [00:20] Could someone perhaps take a look at http://revu.tauware.de/details.py?package=hardware-connected? (A small program that checks whether given hardware exists in the system; suitable for scripting) [00:20] Ubulette: it will go into universe per default. An MIR for main promotion should turn up no issues, but it's up to the Ub moz team to request that. [00:21] ok, i'll ask asac then [00:22] i'm part of ub moz team btw ;) [00:22] (yes, I know :-) [00:25] heh [00:25] /* Why yes, I do have the NIH syndrome. */ [00:25] eh? [00:26] referring to ion_ :-) [00:26] oh [00:28] Hey guy's i've reuploaded my urbanterror and urbanterror-data packages, fixing most if not all the issues persia brought up on them. Could someone review them if possible? [00:28] you're queued (save -data; it's too large for me to download without having to resort to MAC-changing hacks) [00:30] ion_: not a blocker, but consider putting redistribution terms in your man pages, too. Never know who's going to review it for, say, Debian, and might be more pedantic than me. [00:32] Shall i do it immediately, or for the next release? [00:32] wouldn't hurt for the former [00:33] Ok, i’ll upload it in a bit. [00:33] I wouldn't -not- advocate it simply because of them being missing, however. [00:34] crimsun were you referring to me with "you're queued..." [00:34] jscinoz: yes. I'm processing another source package currently. [00:35] oh :) [00:37] I don’t really get how GPL-2 affects documentation (whereas i do understand its implications on *code), but i guess i’ll just throw it to the man pages as well. :-P [00:38] I mean, it’s not as if anyone’s going to create a new product from the man page alone. :-) [00:54] crimsun: Ok, i uploaded a new package with licensing info in the man page. It should be visible on REVU in six minutes or so. [00:59] hi [01:00] i was wondering, i plan to update a new upstream version of package [01:00] is it better to do it debian or ubuntu side ? [01:00] former if at all possible [01:01] which one to start with ? [01:02] rzr: do the work in the Debian repo, then request an Ubuntu sync from Debian [01:02] ok that's what i thought too but it may be slow [01:03] I've created a set of (Prism) webapps for google maps, with fullscreen and auto-hidden control. [01:03] now I want to package that [01:03] Does this make sense: http://paste.ubuntu.com/3184/ ?? [01:05] homepage goes under source i believe [01:05] not here, I want homepages per webapp === asac_ is now known as asac [01:06] oh right [01:06] i've already packaged prism and some webapps so the syntax is correct. Here, I'm not sure about fr and de [01:08] i looooove google maps [01:08] i can spend hours on it [01:09] rzr: you can e-mail the maintainers and utilise #debian-mentors on OFTC. [01:09] i am in contact w/ them [01:09] will try then [01:30] hey guys I want to gt back to motu work tomorrow [01:30] any areas where I should focus... or should I jst try bug fixes... [01:30] ? [01:33] effie_jayx: bug fixes would be great. There's also still a large number of packages unique to Ubuntu and missing watch files. Also, it's coming near time to start testing upgrade paths: if something in Dapper or Gutsy doesn't upgrade to hardy cleanly, this should be fixed. And there is always the FTBFS queue. [01:35] good advices as always [01:36] persia, great [01:37] persia, did you read my question about google maps / prism ? [01:38] Ubulette: Yes. Is this just a selection of prism shortcuts? === asac_ is now known as asac [01:38] persia, let me see if I can decode all you said ;)... packages unique to ubuntu (merges) ? [01:39] effie_jayx: No. packages unique to Ubuntu aren't merges, because they aren't maintained in Debian. See http://qa.ubuntuwire.com/uehs/ for the lists of missing watch files, broken watch files, and packages not in sync with upstream. [01:39] I can help with watch files [01:39] Vorian: We're missing 460. Please do :) [01:40] persia, i've created some css to customize the webapps, that's the added value compared to the -classic one [01:40] persia: I'm on it :) [01:40] Vorian: great. [01:40] persia: what's the procedure after the watch file is added? [01:40] persia, k I shall focus on them and see If I can be more productive... [01:41] Ubulette: OK. I can now see the rationale for the package. Next question: why not have a prism-plugins package that contains all the various services you might want to add later, to save common data, rather than prism-$myapp packages for each interesting set of packages. [01:41] ? [01:41] happy new year! [01:41] hey nxvl, happy new year to you :) [01:42] effie_jayx: OK. Watch files are good, but if you get bored with them, consider looking at upgrade paths and FTBFS (and always, bugs need fixing). [01:42] persia, hmm. those are not plugins at all, but standalone applications.. the difference with any other app is that it's web based, through prism (and xulrunner) [01:43] persia: I uploaded apt-mark-sync 0.0.2-0ubuntu1, as crimsun suggested adding licensing info to man pages as well, but i understand what you wrote in the previous upload’s comments (about waiting for archive admins). [01:43] i am writing this down [01:43] ion_: The issue is that once it is in the NEW queue, nobody is going to upload it from REVU. If it gets rejected, please resubmit. If it gets accepted, a debdiff (or interdiff) addressing the outstanding issues would be most welcome. [01:44] persia, each deb contains at least a .webapp file, a .desktop file and one icon [01:45] Ubulette: How about prism-applications then. I'm just thinking you don't need lots of different debian/rules and debian/copyright, etc. files for each of the web applications. Different binary packages is good, perhaps all depending on a -common and symlinking to /usr/share/docs/prism-applications-common [01:47] * persia wishes LP worked nicely with prism [01:47] persia, to share what ? i don't see what would be in the commom package [01:48] Ubulette: /usr/share/docs/prism-applications/* (copyright, changelog, README, etc.). [01:48] LP, me too, but that mean changing the design (of LP) [01:48] +s [01:49] Maybe nice to have some documentation on extending it too, which might encourage upstreams using Ubuntu to submit bugs for inclusion of their applications (when they work with prism). [01:49] persia, what does copyright mean in that context ? i'm author of the css file, not the icon and of course not of the web site [01:50] Ubulette: /usr/share/doc//copyright is pulled from debian/copyright. [01:50] (in which you'd indicate the copyright owners of the css file, and the icon) [01:51] yes but then, what is shared in common ? it's always different per webapp [01:53] Ubulette: If you have one source package per webapp, you need to store all of a .dsc, a diff.gz, and an orig.tar.gz on the servers, plus a full binary for each architecture for each application. As these are all small, they would compress together with considerable savings, so the total disk space used by the archives would be smaller. A minor point (and per-app isn't wrong), but maybe interesting. [01:53] By putting them all in a single source package, you only have one debian/copyright (which goes in -common), one debian/rules, etc. [01:56] hm. i see. something like prism-webapps as unique source pkg, with dirs grouping webapps per family, one being google-maps (ie my current source tree) [01:57] and 1 bin deb per webapp [01:57] Ubulette: Right. As an added benefit, you wouldn't need to go through source NEW for each new webapp (although you still have binary NEW). [01:57] Ubulette: Unless there's a strong argument against (say a main/universe split, etc.), you might just stuff all the new binaries into the prism source, if you like. [01:58] Actually, I take that back. The prism source tracks upstream, and prism-webapps would have a different upstream (some LP branch). [01:58] i want to keep prism alone. [01:58] yep [02:00] hmm, i feel there's something good here. thanks [02:00] about doing .fr .de .co.jp .whatever, is that a problem ? [02:01] it's not really localization [02:01] Doesn't google maps mirror it correctly? If not, you'd get style points by picking a mirror based either on user-locale or user-physical-location (assuming you have a reliable means to determine this). [02:02] (so a single app that checked the current status, and called the right prism shortcut) [02:07] it's not really a mirror. maps are probably the same but services are not always in sync, and languages are different for everything. seems difficult to auto detect that.. [02:08] Ubulette: The user locale should at least give you a preferred language. It's not wrong to have separate binaries, but matches "just works" better if users are automatically redirected to their preferred language environment. [02:09] my desktop is US, yet I'm french, and I read/type jp and cn sometimes. which webapp would I want? [02:09] presumeably what your locale is set to [02:09] thats the reason for it [02:09] what is the watchfile syntax for git snapshots [02:10] Vorian: Does upstream never release at all? [02:10] my bad [02:11] thanks for the reminder :) [02:11] Vorian: Sure. Just track the latest release (and the snapshot should be newer). When upstream makes the next release, it should tell us that we don't need to track a snapshot anymore (which is good). [02:11] (I'll do the watch files for mozilla packages, I wanted that to be part of mozilla-devscripts sometimes) [02:12] Ubulette: watch files must be in the affected source. On the other hand, having mozilla-devscripts use the watch files to keep the team informed sounds like a good way to do things. [02:12] persia, that's what i meant. [02:13] Excellent then :) [02:13] -be part of+be used by [02:20] can you suggest me a page to read to sync a package like https://bugs.launchpad.net/ubuntu/+bug/177986 ? [02:20] Launchpad bug 177986 in ubuntu "please sync package whitedune from debian experimental" [Undecided,New] [02:21] rzr: https://wiki.ubuntu.com/SyncRequestProcess [02:21] ok perfect [02:22] rzr: We've passed DebianImportFreeze, so you'll want to include a short Rationale explaining why the sync deserves a freeze exception (many do, but to make it easier for the person approving the freeze exception) [02:23] actually i can wait next freeze [02:23] rzr: Would the sync you wanted to request not benefit hardy? [02:24] no, i am not hurry [02:35] persia, initially, I wanted this google-maps thing to be native (as it was so small). I bet you're opposed to the idea, right? [02:36] Ubulette: Yep. Especially so as other distributions that ship prism would likely also benefit from some nice CSS wrappers to make certain applications look nicer. [02:37] yep [02:38] i wish some css artists could make google-reader and gmail looks more gnomish [02:38] prism is gnome only now? [02:38] no [02:38] ahh then very bad idea [02:39] lol [02:39] imbrandon: Why? If there were selectable CSS to match shipped desktop themes, wouldn't that be an improvement? [02:40] sure if it was selectible [02:41] but as just discovered with the maps localization , apparently prism apps cant choose css at rumtime [02:41] imbrandon: Right. So let's welcome CSS submissions for Ubuntu, Kubuntu, Xubuntu, etc. and then complain if it doesn't work right (it would be a bug). [02:41] * persia advocates wrapper scripts [02:42] well with wrapper scripts you have more than one binary, i'd rather think of it as a bug in prism that cant select css at runtime [02:42] imbrandon: That works for me. File a bug. [02:43] after all the css is a conffile in this case and most apps accept a conffile location overide [02:43] on cli [02:43] * persia doesn't think css should be a conffile, only a configuration file. [02:44] changing css at runtime is possible in firefox so as it's also a xul app, prism should be able to do it at some point (it does not now) === bmk789 is now known as bmk789_brb [02:46] google reader is good for languages, you can just change it at runtime in the settings [02:46] persia, have you had a chance to review my changed urbanterror packages? [02:47] jscinoz: No, and I probably won't be reviewing much again until next Monday (especially for large, complex packages). I suspect you've fixed most of what I found, and think someone else would likely give you better feedback. [02:48] ok cheers :) [02:49] also for the old server source package which is no longer needed since both source packages are combined into one, do i need to do something to remove it or just make a comment saying that and leave it? [02:49] I already archived it, based on your advertisement for the new package earlier. [02:49] thanks :) [02:55] mmmm if my meta-data was free i would be happy ... [02:55] * imbrandon pondersa moment [02:56] anything on revu need doing before i get off in my own little world ? [02:56] ugh i have to return to work tomarrow after this long holiday :( [02:58] hum. now, i didn't mean to upload mplayer [02:58] Hobbsee: heh === bmk789_brb is now known as bmk789 === bigon is now known as bigon` [04:00] happy New Year MOTU Land! [04:00] LaserJock: !!!! :D [04:00] Happy New Year LaserJock [04:00] I'm finally back home [04:01] hey LaserJock! [04:01] LaserJock: ponies! [04:01] ... with internet [04:01] LaserJock: so, now you have time to put up the ponies, which you carefully wrote while you were away [04:01] Hobbsee: ah yeah, I was trying to work on it some over the vacation but the lack of internet was difficult [04:02] PONIES! [04:02] my parents 28.8 dialup was ... painful [04:03] Where do they live, the Sahara? [04:03] somewhere in australia [04:03] no [04:03] * StevenK scoffs [04:03] Montana [04:03] they are only 5 miles from town [04:03] hey now, I'm on a 28.8 ATM. [04:03] but they can't even get full dialup speeds [04:04] LaserJock: Is their phone line pair gained? [04:04] LaserJock! [04:04] they are looking into doing internet over satellite [04:04] but it's pretty pricey [04:04] hi imbrandon [04:05] Internet over satellite is expensive and laggy [04:05] I managed to snag myself a shiny new laptop over Christmas [04:05] nice [04:05] StevenK: yep, but that's pretty much their only option [04:05] One-way is ~ 400ms, two-way is ~700ms [04:05] SSH over two-way satellite is somewhat painful [04:08] this new laptop has a big 12cell battery, I've only heard of 6 and 9cell [04:08] yea i had to use satalite in TX , was not fun, at times dialup is less laggy [04:08] but it gets ~ 5+hrs right now so that's cool [04:08] LaserJock: nice [04:09] man i'm gonna go nuts no smokin, ugh [04:09] imbrandon: heh, I saw that [04:09] * StevenK waits for imbrandon to start holding his pen like a cigarette [04:09] lol [04:10] "Trying to quit, are you?" 'Yeah, how can you tell?' "You've been trying to light your pen for three minutes" [04:10] BWHAHAHA [04:10] sounds about right [04:10] i've started chewing toothpicks, i have already went through one whole box in 24 hours [04:11] not even 24 hours [04:11] I'm not sure if splinters are any better [04:11] heh, my trashcan by my desk is full og mtdew cans and splintered toothpicks [04:11] of* [04:11] lol [04:12] imbrandon: how's MOTU SRU going? [04:12] LaserJock: good, not much action atm but seems to be working out well [04:12] i'd imagine mid jan things to pickup [04:13] I'm still going through email but it looks like we might need to clarify/document some stuff on -motu ML [04:13] i've given up on the flashplugin thing, its a no win any way i have gone about it, i'm for removal at this point [04:13] LaserJock: yea, a little clarification might be in order but all in all i think its ok [04:14] mostly it was just those not on irc that got confused , but a simple email should clear things up [04:15] mhm [04:15] we cant even do the 60mb thing with a clear mind, it has security flaws [04:15] so to fix the security issues we have to break konqueror [04:15] and then presumeably fix it [04:16] I’m going not to ever have started smoking in the first place. :-) [04:16] ion_: that would be a good thing [04:17] * StevenK quit smoking years ago [04:17] adobe really screwed us on this one [04:18] (New tenses for time travelers: going not to have ...ed) [04:18] imbrandon: flash? [04:18] yea [04:19] I got sick of being completly out of breath after running up a flight of stairs [04:19] StevenK: exactly [04:20] why does it do that? just too much crap in your lungs? [04:20] StevenK: You just forgot to maintain the extensive level of exercise required to smoke (not that smoking is good, or anything) [04:20] LaserJock: deminished lung capasity [04:22] lucky for my pocket book though toothpicks are far cheaper and have less side effects [04:22] hehe [04:23] yeah, cost alone would make me not want to start [04:23] $0.50 a box , per week VS. $5 a day for cigarettes [04:23] imbrandon: Yes, but you look like an absolute tool buying fourteen packets of toothpicks [04:23] hahaha [04:23] hahaha [04:24] Who said you don’t look like one for buying cigarettes? ;-) [04:24] Or asking the supermarket for a cartoon of toothpicks ... [04:24] i'm sure they will be far easier to put down [04:24] :) [04:25] if there is a package that is going into ubuntu but that is not in debian do we ignoren the XSBC-O-M warnings? [04:25] ignore* [04:25] joejaxx: no it should still have a @ubuntu maintainer addy iirc [04:25] that is interesting [04:26] so should i put myself as the original maintainer? [04:26] yup [04:26] ok [04:26] the @ubuntu.com requirement is archive-wide, regardless of source [04:26] LaserJock: i was just wondering since it was not in debian :P [04:27] so even for native packages? [04:27] ie without with -0 [04:27] how come XSBC-O-M is not automatically added for everything sync'd? (like sometimes when ppl make debdiffs they have to fix XSBC-O-M) [04:27] joejaxx: What native package? [04:27] yea unfortunately the spec just obsoleted the Maintainer Field imho [04:27] joejaxx: yep, just one of the side effects of implementing something archive-wide to "fix" something debian-specific [04:27] joejaxx: yup [04:27] xtknight: It is added in the binary packages, but we don't modify the source packages. [04:28] lol wow that stinks :\ [04:28] * joejaxx puts himself in there twice :P [04:28] j/k [04:28] joejaxx: if it's got an "ubuntu" in the version it needs an ubuntu address [04:29] and native packages [04:29] ( the few of them ) [04:29] * persia notes that ubuntu maintainers with no XSBC-O-M are welcome, as long as the ubuntu maintainer is a small team or specific person, and the package doesn't exist elsewhere to get an original maintainer. [04:29] imbrandon: if they don't have an ubuntu version they won't need it afaik [04:30] LaserJock: There are a few Ubuntu-native packages with Ubuntu native versions (e.g. ubuntu-meta) [04:30] LaserJock: native means there is ONLY a ubuntu version ( i dont mean debian native ) [04:30] so should i just make it native? [04:30] lol [04:30] joejaxx: no [04:30] imbrandon: Some of them have been adopted by debian (who doesn't use -XdebainY amusingly) [04:30] imbrandon: I mean "ubuntu" in the version [04:30] LaserJock: right i do too [04:31] e.g. ubuntu-meta [04:31] joejaxx: The only reason to have a native package is if you are truly convinced it isn't useful anywhere else. There are lots of annoying issues with passing native packages back and forth with Debian, and even worse ones for other distributions. [04:31] well we can have Ubuntu native packages without "ubuntu" in the versino [04:31] persia: oh ok [04:31] LaserJock: right and they still need a @ubuntu addy [04:31] and those wouldn't fall under the address checking [04:31] imbrandon: no they don't [04:31] LaserJock: why not ? [04:32] because there is not policy that says so, as far as I know [04:32] Anyone feel like taking a look at http://revu.tauware.de/details.py?package=hardware-connected (a program to check whether certain hardware is connected; for scripting)? I believe i have taken care of the problems pointed out so far. Thanks. :-) [04:32] joejaxx: Also, a lot of packages that are mostly Ubuntu aren't native. mythbuntu-* is a good set of examples. The packaging is all done in LP, and pulled occasionally into the distro. Nobody else uses it, but they could if they wish. [04:32] sure its the same policy, just because there is ONLY ubuntu changes doesnt make it less a ubuntu change [04:33] imbrandon: no, the debian maintainer spec only says that if it has "ubuntu" in the version that it needs an ubuntu address [04:33] i only brought it up because i put the version as -0ubuntu1 and dpkg-source was complaining :P [04:33] joejaxx: rightfully so [04:33] LaserJock: The debian maintainer spec only applies to packages maintained in Debian. [04:33] persia: not exactly [04:33] LaserJock: ok, thats a bug imho, but ok [04:33] it's enforced archive-wide [04:33] joejaxx: For non-native, you need to have a separate orig.tar.gz [04:33] and this is why i am confused [04:33] as i am not a debian maintainer [04:34] LaserJock: Right. That's just pragmatism. [04:34] joejaxx: You don't need to have it, but people like to see it in case they have a question about the initial packaging later. In my opinion, it's largely useless (as long as Maintainer is an Ubuntu address). [04:35] persia: right, but that's the "policy" we have [04:35] what the archive will and will not accept [04:35] anyway, gotta run [04:35] l8tr [04:35] LaserJock: The policy we have is that all packages need to have an ubuntu maintainer. We don't have a policy that all packages must have XSBC-Original-Maintainer, although many people use that. [04:35] first day back at work tomorrow [04:35] i do not even know if i have a ubuntu email yet [04:35] lol [04:35] or where it points to [04:36] if i do have one [04:36] it points to your LP prefered mail [04:36] imbrandon: is that automatic? [04:36] persia: the spec states "If a source package is modified relative to Debian (this can be determined automatically by examining the version number), its Maintainer field should be updated either as above, or with a more appropriate Ubuntu contact if one exists." [04:36] joejaxx: semi automatic, run once or twice a week by elmo [04:37] oh ok [04:37] LaserJock: Right. "relative to Debian". [04:37] so clearly an Ubuntu native package doesn't fall under that [04:37] LaserJock: -0ubuntu1 means not in debian but will/could be someday, still relitice [04:37] relitive [04:37] The argument would be that all our Ubuntu-unique packages are numbered incorrectly, but that's a different issue. [04:37] imbrandon: right, but you don't have to have "ubuntu" in the version [04:37] uh oh i hope pbuilder does not error out :\ [04:37] LaserJock: Right. [04:38] our versioning is messed up a lot of times [04:38] LaserJock: sure , the archive wont accept a 1.0-0 versioned package, thats against policy too [04:38] imbrandon: no, but I have native packages that have no "ubuntu" [04:38] imbrandon: Sure it does, 1.0-0 is valid [04:38] StevenK: valid for dpkg , not according to the package guide though [04:39] there's nothing that policy that states that a package originating in Ubuntu must have "ubuntu" in the version [04:39] LaserJock: true, but the Maintainer spec should imho cover ubuntu native packages too [04:39] since they are native someone in ubuntu must maintain them [04:40] if "ubuntu" is in the version or not [04:40] so just put motu as the maintainer and me as original and that is it right ? :) [04:40] imbrandon: Someone needs to maintain them. If the maintainer is MOTU, we prefer XSBC-O-M, but it's not a policy requirement. [04:40] persia: right [04:40] joejaxx: correct [04:41] joejaxx: whats your LP id ? [04:41] lp id? [04:41] lp user name [04:41] joejaxx [04:43] great pbuilder create did not fail [04:43] /topic [04:43] Um... right. [04:44] Happy new year, MOTU :) [04:44] joejaxx: check your @@fluxbuntu.org email, i just sent a test message to your @ubuntu.com addy [04:45] nope [04:45] i did not receive it === bmk789 is now known as bmk789_sleep [04:45] give it a few minutes, my @{k}ubuntu.{org,com} email is normaly slow [04:45] oh ok [04:46] i dident get a bounce , so it should go "somewhere" [04:47] up there it is [04:47] there you go :) [04:47] :) [04:47] apparently it works , heh [04:48] iirc you have to request if you want something other than @ubuntu.com , e.g. @kubuntu.org or @xubuntu.org [04:48] but elmo is normaly happy to accomidate [04:49] oh ok :) [04:52] ugh, my dad ( hardcore windows guy ) just sent me a link to Freespire [04:52] asking "have i seen this yet" [04:52] * imbrandon faints [04:52] * Hobbsee steals all imbrandon's smokes, then [04:52] i guess some linux is better than no linux though [04:53] Hobbsee: hehe i made sure they were all gone before new years [04:53] just so i wouldent be "tempted" [04:53] leaste now if i give up i have to goto the store , no one else in the house smokes [05:03] * imbrandon just not realized that sse3 in /proc/cpuinfo == pni [05:03] s/not/now [05:05] :) [05:56] anyone have a link to the repository rules for debian? [05:56] joejaxx: What do you mean by "repository rules"? [05:57] as in section migration [05:57] experimental -> unstable -> testing [05:57] etc [05:57] * StevenK tries to debug rails not working by strace'ing apache [05:57] exirmential -> unstable is manual iirc, no automaitc at all [05:58] Ah. experimental -> unstable is manual, by maintainer action. unstable -> testing depends on urgency and the presence of new bugs. [05:58] unstable --> testing , no new RC bugs and 10 days old iirc [05:58] StevenK: Consider using a different front-end, and only using apache as a proxy server. Rails is known to act oddly with Apache. [05:58] imbrandon: That matches my observations. [05:59] imbrandon: persia yeah but i was wondering if that was posted somewhere [05:59] :D [05:59] persia: I am, though [05:59] * joejaxx is building cryptofs [06:00] :P [06:00] StevenK: Ah. I never thought stracing the proxy server would be more informative than watching tcp traffic, but perhaps it's just me. I suspect your routes. [06:00] * joejaxx wishes there were more stenographic filesystems :( [06:01] * RAOF wonders why joejaxx is interested in stenographic filesystems. [06:01] all the current ones lack development and/or are for 2.{2,4} [06:01] persia: Mongrel works if I connect to it directly [06:01] RAOF: :) [06:02] StevenK: Ah. So it's not actually one of the many ways in which Rails doesn't work :) [06:02] joejaxx: I mean; why stenographic? [06:02] persia: I have a rewrite rule to throw stuff to the cluster of mongrels, too. [06:03] persia: It seems Apache is picking dispatch.cgi on it's own [06:03] joejaxx: you need to check the Britnet docs probably [06:03] joejaxx: [06:03] Britney starts considering packages after 10, 5 or 2 days, depending on how [06:03] urgent the uploads since the last testing migration are. A package needs to [06:03] be in sync on all architectures, and also be less buggy than the package [06:03] currently in testing. Also, they must not break another package in testing. [06:03] Britney is a python script, mixed with c-code, and available at [06:03] http://ftp-master.debian.org/testing/update_out_code/ [06:03] joejaxx: What's the point? You'd need a massive library of known clean data to make it worthwhile. Stenography is interesting because it's hidden, so stenography against single files is easier to hide than stenography against a collection (although I suppose you could download lots of kiddieporn, and nobody would bother digging up your secret stuff while complaining about it) [06:05] RAOF haha sorry i spelled it wrong [06:05] StevenK: Very odd. I always used mod_proxy (but may now be able to safely forget about Rails again). [06:05] steganographic [06:05] sorry there bag typo [06:05] :P [06:05] sjdfsljdfslkd [06:06] bad* [06:06] persia: I've set up a Proxy balancer, and have a bunch of rewrite rules [06:06] * persia fails completely at being sufficiently pedantic [06:06] grr want my thinklight back :( [06:06] imbrandon: thanks [06:06] StevenK: And it still reaches dispatch.cgi? Odd. Some people use pound to avoid some of that. [06:07] Impressive. Now I get 403 [06:07] RAOF: i do not want a shorthand filesystem :P [06:07] LOL [06:09]