Answer
stringlengths
12
9.02k
Id
stringlengths
1
5
CreationDate
stringlengths
23
23
Tags
stringlengths
5
66
Body
stringlengths
40
11.6k
Title
stringlengths
15
140
<p>The next build will allow new users to edit their own posts in this case.</p> <p>Essentially, we disable the image/anchor restrictions for new users once the post already contains the requisite content.</p>
989
2012-02-17T15:58:19.110
|bug|status-completed|new-users|
<p>This bug report was brought on by this question: <a href="https://electronics.stackexchange.com/questions/26704/help-identify-a-motor-and-how-would-i-control-it-with-an-arduino">Help identifying a motor and how to control it with an Arduino</a>. Here is its <a href="https://electronics.stackexchange.com/posts/26704/timeline">timeline</a>. The summary is that a new user attempted to post a question with some images. </p> <p>This, of course, is not allowed due to previous problems with the posting of...er...<em>unsavory</em> content by new users. The <a href="https://electronics.stackexchange.com/privileges/new-user">New User Restrictions</a> state that new users cannot:</p> <ul> <li>post images</li> <li>post more than two hyperlinks at a time, either in a post or their user profile</li> <li>contribute answers to protected questions</li> <li>ask or answer questions too rapidly </li> </ul> <p>This all makes sense. In an attempt to work around this problem, the user uploaded some images to the web and posted two hyperlinks to allow viewers to open those images. This is a very normal response for a new user.</p> <p>Unfortunately, the user neglected to use the "shift" key throughout the entirety of the post. This is also a very normal response. I attempted to correct the illegible question and unprofessional post by inserting the images in Revision 2 and posting a comment requesting that proper capitalization be used. </p> <p>Less than an hour later, the user responded by editing the post and making the desired corrections. Fantastic, right? We've now got images inserted by a 10k mod (so those should be fine) and some changes to the text of the question by the OP. Everything is working as desired - until this showed up:</p> <p><img src="https://i.stack.imgur.com/aoWtb.png" alt="enter image description here"></p> <p>The new user didn't post an image. They didn't touch the lines in which the image uploading code was posted. <strong>New users should be allowed to respond to feedback on their post by editing the question, even if additional hyperlinks or images have been inserted by other users.</strong></p> <p>The check for whether or not the new user has violated the new user restriction should exclude text added by users who are past the new user restrictions.</p>
New users can't edit posts and leave images edited in by privileged users
<p>8 - It encourages sloppy communication. I can just imagine <i>"Um, this resistor thing, like, um, is connected to the third leg of the transistor, um, ..."</i>. No thanks. Forcing people to sit down and write up the relevant parts of the problem using words, diagrams, and pictures is a good thing. We get enough muddled questions as it is, let's not invite a whole new level of messiness.</p> <p>9 - It's hard enough to follow the written word of some people who don't know english well. Having to understand words thru a thick accent will make it even harder.</p> <p>&nbsp;</p>
1012
2012-03-01T00:50:53.907
|discussion|feature-request|
<p>Allowing users to add a "voice recording " that explains their problem ? </p> <p>In my humble opinion it will be great thing if it happens . </p>
Allowing users to add a "voice recording " that explains their problem?
<p>I like the question. The OP is clearly approaching the problem from an engineering point of view, and wants to know not only how but why a solution will work. Referencing the ARRL Handbook, baluns, ribbon cable, and the difference between "balanced" and "unbalanced" lines isn't something the common consumer will do. This question addresses a common problem in radio transmission.</p> <p>Questions that garner quality on-topic answers should stay on this site (and get upvotes!). Questions that don't warrant a quality answer aren't any use to most people, and this includes overly simplistic questions. Where is the line between quality and crap answers? I'll let you know, by voting ;) ...</p>
1024
2012-03-10T05:15:12.847
|discussion|
<p>Referring specifically to question <a href="https://electronics.stackexchange.com/questions/27811/transmission-line-for-uhf-tv-antenna">Transmission line for UHF TV antenna</a></p> <p>I don't see any problem with the question, but apparently there is possibly some disagreement. I took it as a which part is better for the job kind of question, similar to "which is better for high speed logic chips, TTL or CMOS" or "are vacuum tubes or transistors better at surviving EMP".</p> <p>Am I looking at the question in the proper way or am I missing something? Further is there any reason why such questions shouldn't be asked on the forum?</p>
What's the issue (if any) asking about what part is better for the job?
<p>This happened cause this post: <a href="https://electronics.stackexchange.com/a/28255/1833">https://electronics.stackexchange.com/a/28255/1833</a> crashed our diff engine. </p> <p>It crashed our Markdown engine cause we hit some sort of internal edge case, the body of the post has some irregular html, this is causing the Markdown engine to totally collapse. </p> <p>Apologies, I rejected the edit, can the community (somebody with full edit rights) edit that post so it is valid markdown (kill all the tags). </p> <p>We are also looking at fixing the underlying Markdown engine. </p>
1037
2012-03-19T12:18:05.910
|bug|status-completed|
<p>For the last few days (not sure of exactly how many) I have been unable to review any suggested edits. Currently the little bubble is showing me that there are 4, but when I click on this button it tries to go to <a href="https://electronics.stackexchange.com/review/suggested-edits">https://electronics.stackexchange.com/review/suggested-edits</a> but then gives me the error:</p> <blockquote> <p>504 Gateway Time-out</p> <p>The server didn't respond in time.</p> </blockquote> <p>This is happening at home, at work, and on my phone.</p> <p>On my phone I get a slightly different error message, something about my browser not responding in time. I suspect it is supposed to be the same 504 error though.</p>
Can not get to suggested edits review page
<p>It's an April Fool's joke. See <a href="https://meta.stackexchange.com/questions/127701/how-can-i-get-the-april-fools-joke-to-appear">How can I get the April Fool's joke to appear?</a> on the main meta (there's a picture of it in the answer)</p>
1056
2012-04-01T08:11:18.440
|bug|status-bydesign|
<p>When browsing the EE SE I notice a paperclip unicorn come up and ask me whether I want some help with searching through the posts.</p> <p>Ahg, and now it just popped up and said "It looks like you're asking about me. That's nice. - Go on asking - Don't bother me again"!!</p> <p>Is this a virus that I've contracted or is this an April Fool's joke on your site? If it's the former, I would urge you to delete it immediately. If it's the latter, I hope it's not something too serious.</p> <p>I'm using Google Chrome 17.0.963.83 m.</p>
Am I the only one seeing a paperclip unicorn?
<p>We may have miss-communicated some. You can suggest all day long, I have no issue with it, even promote it, but your comment received upvotes and I had received flags. I wanted to post my view and suggest that if you felt the question was strongly off topic to suggest it on meta.</p> <p>Please, do take the time to guide members in ways they can improve the answers they receive. I was stating that I would not take mod action to move the question so that others would hopefully not continue flagging their question.</p>
1061
2012-04-08T01:36:20.573
|discussion|
<p>Kortuk commented:</p> <blockquote> <p>Communication theory has always been considered acceptable here. This is definitely a boundary question, but I think one that is still on topic. If you would like to continue discussion of it @JonnyBoats ask on meta. Right now as a moderator I am not going to take action as I support this as on topic, at its heart all communication theory is applied statistics. – Kortuk♦ 8 hours ago</p> </blockquote> <p>I response to my comment that stats.stacmexchange "may be a better place for this question."</p> <p>I had not flagged or voted to close the question, simply posted the comment. It was not trying to imply that the question was off topic; merely that it <em>might</em> get a good answer on the other site.</p> <p>My primary reason for posting here on meta is to ask when it is appropriate to suggest another stack site? In particular now that there are so many sites it seems that there is often an overlap so that a question could be on topic for more than one site.</p>
REF: Communications engineering, call arrival rate, poisson
<p>Once you open the unanswered questions list, you have to click on <a href="https://electronics.stackexchange.com/unanswered/tagged/?tab=noanswers">"no answers" tab</a> to see questions which don't have any answers.</p>
1070
2012-04-12T11:02:40.520
|support|
<p>I had a look at the <a href="https://electronics.stackexchange.com/unanswered">unanswered question</a>, but I noticed half of them seem to have an answer after all, some even two. Why are they listed there?</p>
unanswered questions
<p>Take a closer look at the <a href="https://electronics.stackexchange.com/faq#bounty">FAQ</a>.</p> <blockquote> <p>In any case, you will <em>always</em> give up the amount of reputation specified in the bounty, so if you start a bounty, be sure to follow up and award your bounty to the best answer!</p> </blockquote>
1072
2012-04-12T15:45:48.673
|support|bounty|
<p>I asked a question, to which I only received somewhat negative comments which questioned my reasoning (and seemingly my sanity for even asking such a question).</p> <p>I felt it was still a valid question, as it definitely relates to electronics, so I offered a bounty. The bounty expired several days ago and there are still no answers. My bounty has been subtracted from my rep but has not been returned.</p> <p>I can't seem to find any information about this particular case in the FAQ.</p> <p>Will I ever get my bounty back?</p>
Where did my bounty go?
<p>If a user account is deleted, that users' up and down votes are removed per <a href="https://meta.stackexchange.com/a/126471">https://meta.stackexchange.com/a/126471</a>. This has always been the case, but it's only recently been reported as 'user was removed'.</p> <p>This can occur when, as suggested by <a href="https://electronics.meta.stackexchange.com/a/1079">Federico</a>, a spam user is destroyed. </p> <p>It's also possible for users to request self-deletion by flagging one of their posts or emailing the team. That's what happened here. The SE team honored the deletion request, the user was deleted, and the user's votes were removed.</p>
1078
2012-04-15T22:41:39.303
|support|
<p>I am curious what does the entry "user was removed" mean in your reputation summary for the day? Here is what I see:</p> <p><img src="https://i.stack.imgur.com/6lDPy.jpg" alt=""></p> <p>See the second line down. Does that mean a whole user account that voted +1 on one of my answers was deleted? Nothing on that line is clickable though, so how can one find out what really happened. If someone got bumped off the site, nosy people like me want to know the jucy details of what they did.</p>
What causes "user was removed" message?
<p>Nope, this isn't about a post on serial communications. As clabacchio noted, this is about reversing votes posted by a single user to another. This is considered fraudulent voting, and has been removed. The votes were on posts which you'd previously received reputation for. Rather than deleting those entries in the log, the effect of the votes were reversed by this entry. The votes were deleted from your posts, but I don't know which posts were affected. A similar thing would happen if someone were to decide that they don't like you and downvote all of your posts.</p> <p>The problem here is that a vote total ought to be a measure of the quality of each question and answer individually based on their content as well as in concert with the author's total reputation. Your accumulated rep is an indicator of your trustworthiness, but each post you make must be able to stand on its own: We won't let you rest on your laurels! </p> <p>What happened here was that someone opened your profile, opened 15 of your answers, and upvoted all of them without voting on any other posts. Take a look at your log from yesterday:</p> <p><img src="https://i.stack.imgur.com/HOUM1.png" alt="enter image description here"></p> <p>Those votes are all from the same user.</p> <p>In some cases, this is evidence of rep-farming: Creating multiple accounts and upvoting yourself. However, I've done some analysis using a combination of common sense about your character and behavior together with some fancy-shmancy tools that analyze your activity on the site and concluded that you're not doing that. Not a difficult conclusion, of course, but it needed to be investigated. </p> <p>Here's a snippet from one of the tools I used showing what happened:</p> <p><img src="https://i.stack.imgur.com/P8nC9.png" alt="enter image description here"></p> <p>Each chord segment in this circle is a user. The chord lengths reflect the site rep of each user. You're the large blue chord at the bottom. I've clicked on the purple user at the upper left which is the serial upvoter. This causes the user chord to radiate arcs of upvotes. The purple arc from this user to you at the left is to scale with the widths of the other arcs terminating at other users: <strong>Of the votes that this user has made, 85% have been to you.</strong> That's not an analysis of posts based on merit, that's a fan club. It distorts the distribution of votes and doesn't help anyone.</p> <p>I've sent a gentle, instructive email to the user referencing this post, so it should stop. However, this could happen again from someone else.</p>
1109
2012-05-07T12:33:31.443
|support|
<p>I got a strange entry in today's rep list called "Serial upvoting reversed". I don't remember a question with that name and was curious to see what it was about, but the really strange thing is that unlike the other entries this one is not clickable. I guess I got a 150 bounty a while back and someone changed their mind who should get it? I know there's been a bunch of discussions about serial ports, although I don't remember a bounty like that specifically. Of course now I'm curious about what I wrote and what the better answer is. How can I see that questions, or is there something else entirely going on? Please explain.</p> <p>Here is what I see:</p> <p><img src="https://i.stack.imgur.com/QKssZ.jpg" alt=""></p>
Serial Upvoting reversed?
<p>Test sizes:</p> <pre><code>$$ \big| \Big| \bigg| \Bigg| $$ </code></pre> <p>$$ \big| \Big| \bigg| \Bigg| $$</p> <p>For some reason, they are not aligned as I was expecting</p> <pre><code>$$ = \bigl| = \\ = \bigm| = \\ = \bigr| = $$ </code></pre> <p>$$ = \bigl| = \\ = \bigm| = \\ = \bigr| = $$</p>
1112
2012-05-07T16:54:36.943
|support|formatting|latex|mathjax|
<p>I tried to write the following equation on the main site using MathJax:</p> <p><img src="https://i.stack.imgur.com/F21Uz.png" alt="enter image description here"></p> <p>The vertical bar and \$a_2=0\$ indicate that this equation should be evaluated at the point where the related parameter \$a_2\$ is zero. It's described further in the <a href="https://electronics.stackexchange.com/a/31390/857">original post</a>. This notation is also used in other mathematical contexts.</p> <p>The traditional LaTeX notation for this symbol is <code>\right|_{a_2=0}</code>, but experimentation shows that <code>\right|</code> is not supported.</p> <p>I've made do in this case by using <code>|_{a_2=0}</code> which produces the following result:</p> <p>\$s_{11} = {\frac{b_1}{a_1}} |_{a_2=0}\$</p> <p>It's legible, but the vertical bar is the wrong height. It should match the height of the fraction, but it's just a normal character and does not stretch. I do note that other characters like <code>(</code> do stretch with <code>\left(</code> and <code>\right)</code>:</p> <p>\$\left(\frac{b_1}{a_1}\right)\$</p> <p>so perhaps I'm just screwing something up. The <a href="http://www.mathjax.org/docs/2.0/tex.html#r" rel="nofollow noreferrer">MathJax documentation</a> says that <code>\right</code> is supported, so I think this should work. What am I doing wrong?</p>
How can I get an 'evaluation at' vertical bar in MathJax? \right| doesn't seem to work, looking for alternatives
<p>The default sort on <code>/users</code> is by reputation, and the default <strong>time frame</strong> is the current month. The following screenshot shows the monthly rep gains by each of the associated users:</p> <p><img src="https://i.stack.imgur.com/2qsOD.png" alt="enter image description here"></p> <p>Additionally, the tags shown are the tags for which the user has earned the most rep in the time frame.</p> <p>Select a time frame of 'all' to view the ordinary rep statistics.</p>
1128
2012-05-13T23:53:32.980
|support|
<p>If you look on the Users page you see a number next to the profile picture. What does this number mean as it doesn't look related to their reputation?</p>
What do the numbers mean on the users page?
<p>As Olin said,</p> <blockquote> <p>Maybe a solution is to not bump a question for small edits</p> </blockquote> <p>Probably the same mechanism has been suggested in meta.SO, but I post it as an answer so we can see better how much agreement it gets. Maybe we can change it!</p>
1132
2012-05-15T08:45:37.490
|discussion|community-wiki|
<p><a href="https://electronics.stackexchange.com/a/31763/2064">This answer of mine</a> was turned in a CW. I'm not sure I'm happy with that. Everybody can edit answers, but as far as I understand it a CW is an <em>invitation</em> for edits. I don't want everybody to mess with my answer. If someone wants to contribute, why doesn't she write her own answer? Answers will be more varied, and can answer per answer be judged for their value, so that the author knows when it needs editing. </p> <p>Now I understand CW won't go away, but shouldn't I at least be consulted before changing my answer to a CW, especially since it's the only answer on the page to which it happened. (It's a different situation if all the page, question + all answers, is made CW.</p> <p>Shouldn't I have been asked first?</p>
Community Wiki (CW) worries
<p>Moderators regularly clear off-topic/not contrustive/rude/out dated comments.</p> <p>Your comments were probably left because they related to adding further detail to the question, ideally those comments are edited into the question as further detail and they are then deleted also. Lets focus on keeping the house tidy. Someone visiting the site will have better results with a clean question and answers, comments are great to get the best answers possible and occasionally there are a few of discussion that are worth keeping but they should be generally included in the related post.</p>
1150
2012-05-25T16:35:37.803
|support|comments|
<p>There was a few (I last remembered at least 5) comments on the question I posted, suddenly disappeared and only left with mine. At least 2 user posted the comments. I seriously doubt they deleted all of it.</p>
Disappeared comments
<p>That page shows a name that looks like the person whom garnered the bounty but it is actually whom last answered/edited the question.</p> <p>You were the last to post an answer, it shows you there.</p>
1184
2012-06-03T08:35:22.780
|support|
<p>Regarding this question: <a href="https://electronics.stackexchange.com/questions/28864/how-to-build-a-2n6027-put">How to build a 2N6027 PUT?</a> which was to rewarded with a +500 bounty.</p> <p>From Kortuk's profile page I understand the bounty was rewarded to me: <a href="https://electronics.stackexchange.com/users/230/kortuk?tab=bounties">https://electronics.stackexchange.com/users/230/kortuk?tab=bounties</a></p> <p>But I am under the impression that I never actually received that bounty. What is the reasoning behind not rewarding the bounty? Did I miss anything or am I misreading the information? Please help me understand the mechanics behind bounties being rewarded or not.</p>
Why was this bounty not rewarded?
<p>Your rep not being high enough to see deleted questions seems to make it so that you cannot see the affects from a deleted question. Here is a heavily cropped picture of your rep changes, I did my best to hide secret moderator buttons and such. </p> <p><img src="https://i.stack.imgur.com/yaUgj.png" alt="Image showing reputation gain previous day due to upvote and reputation loss the day after to a post being removed that was edited"></p> <p>My mouse is highlighting the time of the removal event as a time stamp to give the picture more reference then yesterday and 2 days ago.</p> <p>Sorry for the confusion, just another reason to work for 10k rep!</p>
1190
2012-06-04T12:01:25.967
|bug|status-bydesign|
<p>The day before yesterday I had <strong>2010</strong> rep, I got +5 yesterday, so now I'm at <strong>2013</strong>. Makes sense, right? (I don't see anything about downvotes on my rep page.)</p>
New arithmetic in counting reputation?
<p>It was a spam poster trying to get his link on as many questions as possible. It was not a real answer of any sort, they were deleted and you are not yet high enough rep to see deleted content.</p> <p>Sorry for the confusion.</p>
1192
2012-06-04T19:30:17.967
|support|bug|
<p>I saw a notification in the inbox, informing me about 2 new answers on my <a href="https://electronics.stackexchange.com/questions/31296/adding-third-channel-to-the-pic32-starter-kit-dma-chaining-and-pwm-example/">question</a>. I went to check them, but I can't see them. </p> <p>The quote about the answer is "I experimented with looking at your web site in my cellphone and the format doesnt seem to be cor...", which might not be even related to my question. Possibly the notification was just a glitch in the software?</p> <p>I could not find "tech support" for the site itself anywhere, so I'm posting this here. </p>
I got notification on two new answers, I can't see them
<p>I think <a href="https://electronics.stackexchange.com/questions/33478/device-to-record-phone-calls/33479#33479">this</a> was the question you're referring to, based on the fact that it was asked two hours before and answered one hour before your question here. It was self-deleted. The last question to be deleted before this one was two days ago.</p> <p>They can't disappear without a trace, no, but they're harder to find than just scrolling through <code>/questions</code>. Mods can find them with the <code>deleted:1</code> search option, and can view deleted answers in user profiles, but they don't show up in the ordinary listing.</p>
1209
2012-06-08T09:48:48.750
|support|
<p>A recent question had an answer which couldn't quite convince me. But when I wanted to have a look at it again the question seemed to have disappeared completely. In my activity log no trace of the comment I posted, and in answerer's activity log no sign of his answer. </p> <p>I thought questions could only be deleted when they don't have answers yet?</p>
Can questions which have answers disappear without a trace?
<p>Signatures are not allowed, think of the useless noise added by every user adding an extra one line signature. Your signature is your little name and link in the corner.</p> <p>If you want to thank someone upvote great answers and accept the one that solves the issue. This is the built in site reward. </p>
1249
2012-06-18T19:07:09.620
|discussion|asking-questions|signature|
<p>I've 'signed' several of my questions (across the SE platform) like so:</p> <pre><code>Thanks for any help or insights, I appreciate it! </code></pre> <p>or something akin that, sometimes along with my name. My reasoning behind this is that someone is taking a fairly significant chunk of time to help me out, and I want to let them know I really do appreciate it (this feeling is compounded by the fact that I'm relatively inexperienced, so it's not often I can give back by answering questions). However, more often than not, someone edits this out. Not that I really have any problem with it, I'm just wondering what the reasoning behind it is. </p>
Why is 'signing' your question generally frowned upon?
<blockquote> <p>Tag wikis have much higher requirements to edit than other posts, to edit tag wikis <em>without</em> going through the suggestion stage you need the <a href="https://stackoverflow.com/privileges/trusted-user">Trusted User</a> privilege (requiring 20000 rep).</p> <p>Otherwise, all edits move to the suggestion queue, where a few users with the <a href="https://stackoverflow.com/privileges/approve-tag-wiki-edits">Approve Tag Wiki Edits</a> privilege can work together to edit a wiki.</p> </blockquote> <p><a href="https://meta.stackexchange.com/a/94448/152553">From DMA57361 here</a>.</p> <p>This is because when you edit a single post it bumps to the front page and receives review, on tag wikis once they are approved they do not see review unless someone stumbles upon it. Instead we have a review process built in unless you hit 20k rep.</p>
1256
2012-06-22T16:58:18.483
|support|
<p>I just approved a suggested edit by someone with over 5700 rep. According to <a href="https://electronics.stackexchange.com/privileges/edit">this page</a> you only need 2000 rep to edit. So why was approval needed here?</p>
required rep for editing questions and answers
<p>I think we should eliminate the aluminum tag. We have more descriptive or more accurate tags for both uses. For example, aluminum capacitors could be filed under <a href="https://electronics.stackexchange.com/questions/tagged/electrolytic-capacitors" class="post-tag" title="show questions tagged &#39;electrolytic-capacitors&#39;" rel="tag">electrolytic-capacitors</a> and aluminum cases/boxes should just be under <a href="https://electronics.stackexchange.com/questions/tagged/enclosures" class="post-tag" title="show questions tagged &#39;enclosures&#39;" rel="tag">enclosures</a> or <a href="https://electronics.stackexchange.com/questions/tagged/cases" class="post-tag" title="show questions tagged &#39;cases&#39;" rel="tag">cases</a>.</p>
1259
2012-06-23T02:06:56.783
|discussion|tags|
<p>What should we do with <a href="https://electronics.stackexchange.com/questions/tagged/aluminum" class="post-tag" title="show questions tagged &#39;aluminum&#39;" rel="tag">aluminum</a>? There is currently a tag wiki edit proposed to read</p> <blockquote> <p>Aluminum is one type of capacitors. This tag can also be for cases/boxes, if the question is electronics related.</p> </blockquote> <p>I don't think a tag should have two distinct meanings. Do we seperate it into two tags, or just give it one meaning?</p>
What to do with the Aluminum Tag?
<p>The system automatically removes that that only have one use after a time period, a month to my recollection. No need to worry about it, it is automatic!</p>
1304
2012-07-10T18:09:42.953
|discussion|tags|
<p>Most of the ones on the <a href="https://electronics.stackexchange.com/tags?tab=new">new tag page</a> only have 1 question. Should we remove <a href="https://electronics.stackexchange.com/questions/tagged/linker" class="post-tag" title="show questions tagged &#39;linker&#39;" rel="tag">linker</a> <a href="https://electronics.stackexchange.com/questions/tagged/stack" class="post-tag" title="show questions tagged &#39;stack&#39;" rel="tag">stack</a> <a href="https://electronics.stackexchange.com/questions/tagged/ue" class="post-tag" title="show questions tagged &#39;ue&#39;" rel="tag">ue</a>, probably a typo <a href="https://electronics.stackexchange.com/questions/tagged/constants" class="post-tag" title="show questions tagged &#39;constants&#39;" rel="tag">constants</a></p> <p>I already removed <a href="https://electronics.stackexchange.com/questions/tagged/log" class="post-tag" title="show questions tagged &#39;log&#39;" rel="tag">log</a> on the <a href="https://electronics.stackexchange.com/questions/33580/whats-a-good-sensor-for-measuring-high-voltages">one question that used it</a> using the existing datalogger tag. Also removed <a href="https://electronics.stackexchange.com/questions/tagged/nokia" class="post-tag" title="show questions tagged &#39;nokia&#39;" rel="tag">nokia</a> one.</p>
Remove some tags
<p>Flagging is the thing to do. If the flagged comment for instance contains some bad language a mod may edit that out of the comment. If the comment as a whole is inappropriate it may be deleted, as apparently happened here. I know about ever more escalating heated discussions thirty comments long completely deleted. </p> <p>That's one thing. A misbehaving user can also be sanctioned. If there are too many complaints his account may be suspended for a day, a week, a month or a year, IIRC. </p> <p><img src="https://i.stack.imgur.com/5FEyK.png" alt="enter image description here"> </p> <p>So flag, and moderators will do the necessary to keep our site nice.</p>
1318
2012-07-14T19:52:56.003
|discussion|
<p>Sorry if this is the wrong place to have this discussion, but I don't quite know where I should put this. I posted an answer to a question yesterday <a href="https://electronics.stackexchange.com/questions/35663/minimum-operating-temperature-outer-space/35665">Minimum operating temperature - Outer Space?</a> and I noticed when someone else added an answer which I responded to saying I felt it was not answering the question asked.</p> <p>I don't feel that the expanded response is on topic, and the user is now posting comments which contain misinformation on my answer or that address comments posted to his answer. I feel like posting responses which address his comments are only going to exacerbate the problem. What is the appropriate response to "this is on topic but incorrect!" repeatedly?</p> <p>Thanks for any help.</p> <hr> <p><strong>EDIT:</strong> It seems like the outright false comment has been removed, but the question has remains: should I flag comments which are wrong (but on topic) when clearly it's an escalation of an argument I don't think is appropriate for the thread?</p>
Where should we discuss user issues?
<p>Stance on the subject has not changed to my knowledge, diamond moderators act from the same basic set of ground rules as they did before.</p> <h3>A Maturing Community</h3> <p>The community maturing and requiring higher quality posts of its members will really increase the level of shopping question that works. The primary metric for judging the suitability of such questions is described in the <a href="http://blog.stackoverflow.com/2010/09/good-subjective-bad-subjective/">blog post "Good Subjective, Bad Subjective"</a>. Many of the metrics are based on the answers given, as they are often the clearest way to tell if the question is too subjective to be useful. It is still borderline to me but close enough to not be worth the action of a diamond.</p> <h3>Question with Detailed Information but Lack of Research</h3> <p>The first did not show thorough research, the metric by which you are supposed to vote on questions, but did have the needed information to possibly allow users to teach instead of just giving basic links to companies or google. This type of question often spawns from a user not knowing how to effectively search for components, not necessarily a lack of effort on their part. Often showing how to use a tool to look into these parts is helpful to them and users in the future.</p> <h3>Question About Component Selection in a Specific Technical Situation</h3> <p>The second question has a clear technical situation where they need help on how to pick a mosfet. There is an excellent chance for a great answer here really teaching a user how to go about picking a mosfet and looking at its characteristics to see if it can handle the job. Honestly, the point at which this question started getting great answers heavily correlates with the example that one of our top users has set. Ideally you give what I call a stevenvh answer on this site, you have answers that explains step by step what you need and how to tell if your device does it. Stevenvh is not the only user whom does it but his concise communications skills and willingness to attempt a task instead of giving just a part number has been a significant contributor to our site being able to open the flood gates a bit more to these questions.</p> <h3>Raising Flags for Moderators with Low Quality Questions</h3> <p>Flag if something seems like it does not fit or there is an issue. If you think the question is poor and you can, vote to close. Actions on the site are designed to be easily reversible, if a question is closed for a reason, it can be edited and reopened. Moderators often only act if an issue is clear to them. We occasionally err but to my knowledge always correct such situation. Our site allows anyone to post without verification, instead using the community to close crap afterwards. At this point it must be brought to quality guidelines then we will reopen. It allows a shorter delay to good questions showing up and allows us to mitigate the presence of crap on the site.</p>
1332
2012-07-19T23:21:29.570
|discussion|
<p>Recently, I've noticed a bunch of shopping questions with good answers in which no one complained (about them being shopping questions).</p> <p>Most recently, I've seen:</p> <ul> <li><a href="https://electronics.stackexchange.com/questions/36112/help-finding-appropriate-avr-device">Help finding appropriate AVR device</a></li> <li><a href="https://electronics.stackexchange.com/questions/36098/selecting-a-mosfet-for-driving-load-from-logic">Selecting a MOSFET for driving load from logic</a></li> </ul> <p>I've seen more but don't remember the exact titles off the top of my head. Has opinion changed regarding shopping questions? Or are those questions <em>different</em>?</p>
Has our opinion on shopping questions changed?
<p>The next build will update the footer to read "Electrical Engineering"</p>
1335
2012-07-20T13:34:01.190
|discussion|status-completed|
<p>After seeing few consumer electronics questions here today, I thought that it may be a good idea to change the link in the footer of SE pages to something which describes the site better.</p> <p>I think (and I may be wrong) that some users just read "electronics" and assume consumer electronics without reading too much about the site itself. If that is the case, then changing text to something like "electrical engineering" or other suitable name which focuses on the engineering part may decrease the amount of off-topic questions. </p> <p>An established example for a long phrase would be game development. The "development" phrase immediately makes it different from our gaming site. </p> <p><img src="https://i.stack.imgur.com/9fIdl.png" alt="footer"></p>
Should we ask for the link in footer to be changed?
<p>These are either single use tags that will die on their own or no use tags the system has not swept out yet. These will clear out on their own.</p>
1340
2012-07-22T10:27:58.700
|discussion|status-declined|tag-cleanup|
<p>There are currently 141 unused tags on EE, not counting <a href="https://electronics.stackexchange.com/questions/tagged/untagged" class="post-tag" title="show questions tagged &#39;untagged&#39;" rel="tag">untagged</a>. There are irrelevant tags like <a href="https://electronics.stackexchange.com/questions/tagged/australia" class="post-tag" title="show questions tagged &#39;australia&#39;" rel="tag">australia</a>, or mysterious ones like <a href="https://electronics.stackexchange.com/questions/tagged/pe" class="post-tag" title="show questions tagged &#39;pe&#39;" rel="tag">pe</a> (poly-ethylene?). </p> <p>These tags will pop up as suggestions as you start typing part of the word. I think we could discourage their use by deleting them altogether. You would still be able to add them, but then it's a deliberate choice. Now the fact that they appear as suggestion indicates that it's OK to use them, while in many/most cases it isn't, think <a href="https://electronics.stackexchange.com/questions/tagged/australia" class="post-tag" title="show questions tagged &#39;australia&#39;" rel="tag">australia</a>. </p> <p>So I propose to delete them.</p> <p><strong>edit</strong><br> I may be misinterpreting. Other tags show the number of questions between brackets, like "(5 x)" next to them, and these don't. Quick survey says there are indeed some unique appearances, others seem to be there as synonyms. I'm not idiotic enough to check them all. As I understand it the synonyms will be automagically converted, so it's possible that the other ones are indeed used, albeit only once. </p>
What shall we do with the 141 unused tags
<p>On Stack Overflow, there are tags for code libraries and software development platforms. For instance, there's a <a href="https://electronics.stackexchange.com/questions/tagged/jquery" class="post-tag" title="show questions tagged &#39;jquery&#39;" rel="tag">jquery</a> tag. In many cases, it is redundant to list tag synonyms for each release of a library update, so the developers put in place validation to prevent people from creating synonyms like <a href="https://electronics.stackexchange.com/questions/tagged/jquery1.6" class="post-tag" title="show questions tagged &#39;jquery1.6&#39;" rel="tag">jquery1.6</a> and <a href="https://electronics.stackexchange.com/questions/tagged/jquery1.7" class="post-tag" title="show questions tagged &#39;jquery1.7&#39;" rel="tag">jquery1.7</a>.</p> <p>Since this site runs on the same engine, the system thinks you're creating a version-specific tag, and only diamond moderators can create version specific tag synonyms. In short, you're not doing anything wrong, you're just getting caught up in some of the system safeguards.</p>
1396
2012-08-02T06:14:08.087
|support|bug|status-bydesign|
<p>I was working on tags (between answers) and when I wanted to suggest a tag synonym I got this message: </p> <p><img src="https://i.stack.imgur.com/7EMP2.png" alt="enter image description here"></p> <p>What does it mean?</p>
What does this message mean?
<p>Most of our source code questions don't specify a programming language in the tags. I'm not sure if we want every question that has code in it to be tagged with a language, but there is a simple workaround - you can add a header to force a language.</p> <p>For reference, see the <a href="https://electronics.stackexchange.com/editing-help#syntax-highlighting">syntax highlighting for code</a> segment of the advanced editing help guide. The only language that we use that isn't supported is Verilog.</p>
1398
2012-08-02T14:05:13.280
|support|
<p>Syntax highlighting! Mike asked <a href="https://stackoverflow.com/questions/11770464/pic16f-sspadd-causing-scl-to-go-low-early">this question</a> on SO, and the code looks like this: </p> <p><img src="https://i.stack.imgur.com/0c9GH.png" alt="enter image description here"></p> <p>Then he <a href="https://electronics.stackexchange.com/questions/37052/pic16f-sspadd-causing-scl-to-go-low-early">moved the question to EE</a>, which is sensible, because we're so much smarter than them on SO, but here his code looks like this: </p> <p><img src="https://i.stack.imgur.com/Ycb9J.png" alt="enter image description here"></p> <p>What happened to the syntax highlighting?</p>
What does SO have that we don't? Syntax highlighting
<p>List questions are generally not constructive questions for our site. They can be incomplete, out-of-date, and there can be a lot of noise. They are questions with "no right answer", so they are a bad fit for our site where the goal is to find <em>one</em> right answer. See the <a href="https://electronics.stackexchange.com/faq#dontask">FAQ</a> for more guidance on questions that are not good fits here. </p> <p>This is a generic StackExchange policy, but it is open to discussion in Meta. If you (or anyone) thinks that we should change this policy, then we need to have a public discussion and come to a consensus. Your example from CStheory is permitted on <em>that</em> site because <em>they</em> feel those questions are necessary for their site to function.</p> <p>The "5 Vote Rule" is a method of involving the community in moderation duties, such as closing questions. On borderline questions, a mod might defer their binding vote until more users vote, but this is up to their discretion. Also, keep in mind that while only one moderator's name shows up, that doesn't mean that other moderators can't agree with the action as well, or that there was no other community input. Would you feel better if five <em>mods</em> closed your question? Personally, if I think that a question fits one of the close reasons, I will close it.</p> <p>The other part of your question is subjective (in my opinion) so I'm not planning on voting to re-open.</p>
1405
2012-08-10T16:26:12.393
|discussion|closed-questions|
<p>I would like to inquire about the closure of <a href="https://electronics.stackexchange.com/questions/37513/must-read-electronics-papers">this question</a> which Kortuk, singlehandedly, executed.</p> <p>I find the closure description "closed as not constructive" less than helpful, and I would have appreciated a comment from Kortuk.</p> <p>Also, why isn't the "five close votes rule" applied here, where the question isn't obvious spam, nonsense, obscene, etc. ?</p> <p>Finally, I would like to point out that the Computer Science StackExchange has a <a href="https://cstheory.stackexchange.com/questions/1168/what-papers-should-everyone-read">similar question</a> that was not deemed inappropriate (and was greatly appreciated).</p>
Closed question inquiry
<p>This will <a href="http://chat.stackexchange.com/faq#retention">happen automatically</a> in one of two ways, depending on the previous activity in the room.</p> <ul> <li>If the room seems to have some interesting content (see the above link for a precise definition), it will be <em>frozen</em> after 14 days of silence. This means it's hidden from the regular room list (you have to explicitly "show frozen rooms" to see it), and new messages can't be added.</li> <li>If the room does not really seem to have interesting content, it will be <em>deleted</em> after seven days of silence. Deleting is similar to deletion of questions and answers on the Q&amp;A sites: Only users with more than 10,000 reputation can see it; new messages can't be added.</li> </ul> <p>Manual freezing and deletion is limited to moderators. If you feel strongly that a room should disappear immediately, just find a moderator to do it, but usually you can just wait and the inactive room will disappear eventually.</p>
1416
2012-08-29T11:28:49.143
|support|chat|
<p>5 days ago, as an experiment, I clicked on the link that appeared on a comment thread (the one that says "let us continue this discussion in chat"). </p> <p>The system created a chat room for the thread, and copied comments there. Wonderful. </p> <p>But now the room is useless (I said it was an experiment), so: how can I close/delete/remove/whatever that room? It "disappears" by itself when not used for enough days, or a manual action is required?</p>
How to "close" a chat room
<p>This is dependent on the browser you are using, so no, nothing can be done by the Stack Exchange team (short of not using MathJax).</p> <p>One thing you can do is try switching the renderer that MathJax uses. To do this, right click on the MathJax table/text/formula/etc., roll over Math Settings, and then Math Renderer:</p> <p><img src="https://i.stack.imgur.com/0qrv1.png" alt="enter image description here"></p> <p>I think by default HTML-CSS is used (although in the picture I above I switched to SVG). MathML is only supported by certain browsers, but the one I was using didn't have it so I couldn't post a comparison with it enabled. </p>
1443
2012-10-10T09:59:51.897
|support|
<p>The answer at <a href="https://electronics.stackexchange.com/a/42266/1951">https://electronics.stackexchange.com/a/42266/1951</a> looks like this to me:</p> <p><img src="https://i.stack.imgur.com/tvtpX.png" alt="enter image description here"></p> <p>Normal font rendering at the top, but the MathJax stuff below it is pretty bad. Is it possible to fix this somehow?</p>
Can anything be done about this awful font rendering?
<p>As more and more engineers function in global roles, a little reminder now and then that not everybody across the world writes numbers the same way is a good thing. Best to see it here and now than in a failure mode analysis!</p>
1451
2012-10-11T20:37:48.483
|discussion|formatting|
<p>This question:</p> <p><a href="https://electronics.stackexchange.com/q/42873/2028">Why do many electronics operate on 5 AND 3.3 Volts?</a></p> <p>(Worded "Why do many electronics operate on 5 AND 3,3 Volts?" in case it gets edited.)</p> <p>Recently reminded me of the frustration of working with electronics on an international scope. Some countries use period/full-stop for the decimal mark, others use the comma.</p> <p>I read up a little on this at Wikipedia:</p> <p><a href="http://en.wikipedia.org/wiki/Decimal_mark" rel="nofollow noreferrer">http://en.wikipedia.org/wiki/Decimal_mark</a></p> <p>"International" languages like Ido and Esperanto favor use of the comma for the decimal mark. I was surprised by this, because I find the comma more distracting in numbers when reading with text.</p> <p>In any case, my question/proposal is thus:</p> <p>Does StackExchange (specifically EE) have any preference or standardization when it comes to the decimal mark? Which is ultimately more common? (Obviously I would be biased if I said the period/full-stop seems to be more common, as I am in the US.)</p> <p><strong>Should SE/EE standardize on one or another, so as to not mix formats?</strong></p>
SE network or standardized decimal mark?
<p>Hats are cute, I guess, but to me they seem to be largely functionally redundant with badges.</p> <p>My big complaint is that the additional client-side scripting used to implement them is <em>really</em> slowing down my browser. I had to turn them off.</p>
2495
2012-11-15T05:03:11.030
|discussion|
<p>Last year around Christmas time <a href="http://gaming.stackexchange.com">arQAde</a> had an awesome promotion in which they awarded hats for gravatars for completing basic tasks around the site. If you don't remember it, here is a <a href="http://stackexchange.com/promos/2/hat-dash">link</a> the promo from last year (and <a href="http://blog.gaming.stackexchange.com/2011/12/holiday-2011-hat-dash-the-hattening/" rel="nofollow noreferrer">another!</a>).</p> <p><img src="https://i.stack.imgur.com/QOZfx.jpg" alt="hats"></p> <p>This year, SE is planning to run it for <em>everyone</em> because we all love hats, right? right?</p> <p>But to get this awesome hat promotion we have to <strong>opt in</strong>. That means acting on this meta post, whether that's voting it up, answering in the affirmative, positive comments and/or just directing positive energy this direction. </p> <p>I think this is a good chance at a bit of harmless fun, that could potentially help keep site traffic up through the northern hemisphere winter (when traffic usually dies down a bit for us).</p> <p>HOWEVER: this promotion is <strong>optional</strong> for sites, and/or individual users. If the users of this site <em>do not</em> want this in general feel free to voice that opinion. We can opt out of the promotion. Individual users will be able to opt out as well (they will be provided with an "I hate Hats" link to opt out).</p> <p>Again, please let us know what you think as this is your site not ours.</p> <p>(Majority of content blatantly stolen from <a href="https://bicycles.meta.stackexchange.com/users/76/freiheit">freiheit</a>'s copy of <a href="https://christianity.meta.stackexchange.com/users/49/wax-eagle">waxeagle</a>'s <a href="https://christianity.meta.stackexchange.com/questions/1420/do-you-like-hats">post on meta.christianity</a>.)</p>
Do you like hats?
<p>Comment flags are counted as well, and you have 2 comment flags. I show you having 78 helpful flags, and 2 comment flags.</p>
2502
2012-11-23T22:22:42.400
|discussion|
<p>I noticed that I have the deputy badge, which is 80 helpful flags, yet in my profile it says I've only made 77. So where have I got the extra flags from?</p>
How have I got a badge with only part of the requirements?
<p>You can't accept a comment as an answer. But you can suggest the author to change it in an answer, and then accept it. </p>
2504
2012-11-24T15:19:18.253
|support|comments|unanswered-questions|
<p>I have obtained an answer to my question which is in the comment. How can I mark it as the answer. Please visit</p> <p><a href="https://electronics.stackexchange.com/questions/49537/do-all-phy-ethernet-chips-have-a-hard-coded-mac-address">exchangelink</a></p> <p>WoutervanOoijen has given the correct answer to my question and I want to mark my question as answered but there is no Check(tick) option on the comments.</p>
How can I mark a comment as the right answer?
<p>This should be changed in the FAQ section, under <a href="https://electronics.stackexchange.com/faq#questions">What kind of questions can I ask here</a>, in the NOT to ask part, it says "a shopping or buying recommendation." </p> <p>Because there are plenty of EE questions about components that are valid and it seems like they get down-voted fast, I think it should be reworded to something like, "no vague/effortless shopping questions." Over time this will help reduce inappropriate down votes.</p>
2506
2012-11-24T20:51:11.667
|discussion|
<p>Reading <a href="https://electronics.stackexchange.com/questions/49733/lcd-modules-with-i2c-or-spi-interface">this question</a> got me to thinking... why do we so readily downvote these types of questions? It's apparent that the user doesn't even really have knowledge of whether or not these modules exist... and he's looking to potentially head in a small scale production run.</p> <p>To me, these are some of the things related to shopping that we, as a community, are more than qualified to answer. Something that would otherwise be extremely hard to find on your own. Things like volume pricing for custom modules, etc... that's hard to find information. We have people here who design devices that get mass produced... these guys have sometimes intimate knowledge in purchasing, etc.</p> <p>Can we re-quantify the true meaning of a "shopping" question and maybe show a little more love, and spread a little more knowledge, to dudes like this?</p>
Can we ease up with the negative association to "shopping" questions?
<p>The PCB shouldn't be auto-routed (at least, not excessively) to be eligible for the review here. May be, we should add it to this unofficial policy.</p> <p>Here's a <a href="https://electronics.stackexchange.com/q/72968/7036">thread</a>, which is an unsuccessful test case. Unfortunately, the O.P. there have fully auto-routed his very-first layout. In addition, the schematic wasn't posted. Reviewing that doesn't make a lot of sense. We don't even know what the auto-router settings were.</p>
2513
2012-12-05T06:55:26.573
|discussion|
<p>I think PCB review questions like "Here's my PCB, can anyone tell me if my [layout/vias/placement/whatever] is good?" are not a good fit for the site.</p> <p>That said, are there any sites that do PCB review/consulting work inexpensively or for hobbyists?</p> <p>If not, just as <a href="https://stackoverflow.com/">StackOverflow</a> has <a href="https://codereview.stackexchange.com/">CodeReview</a>, I wonder if anything related to schematic/PCB review would be a worthwhile endeavor, or just a colossal waste of time. Thoughts?</p>
PCB review questions: acceptable? Separate SE site?
<p>This is an interesting case, as the examples of the specific questions might actually lead to rabbit holing the wrong answer. Eg. if you're trying to find a beacon at that frequency in that sort of area you'd probably do better with a doppler direction finder. I think only the over arching question about methods would elicit that response, rather than getting stuck in a series of questions about designing a signal strength meter.</p>
2523
2012-12-16T05:35:24.967
|discussion|asking-questions|design|
<p>Having gone through meta for such question, and finding one regarding circuit design review, I completely understand that view that "circuit designs" dumped on community for crowd-sourcing review comments is bad and frowned upon.</p> <p>However, was wondering if there is some acceptable way, which amounts to asking for help design circuit, that is definitely beyond one's league in terms of understanding / appreciating in detail ?</p> <p>For example, I am (a noob, especially when it comes to UHF RF electronics) trying to create a solution that tries to locate 434MHz RF beacon signal, in an area the size of a small farm. Now I have "found" an almost ready-to-use circuit schematic that can convert RF signal strength dB value (scaled DC output). However it has a relatively wide operating frequency range, which covers GSM (900MHz, 1800MHz), 3G (2100MHz) and WLAN (2400MHz) as well, where there is significant chatter, which is why I need a band-pass filter to help me detect signal strength in 434MHz range. Having searched with terms like "434MHz band pass filter", I have come across <a href="http://www.raltron.com/cust/tools/band_pass_filters.asp" rel="nofollow">this calculator</a>. For the input values, I am using value of RL=53 Ohm, Freq=434 MHz, but not sure what to use for loaded-Q ? From <a href="http://en.wikipedia.org/wiki/Band-pass_filter" rel="nofollow">this wikipedia article</a> on band-pass filters, I think the loaded-Q might be the Q-factor(!), and what I believe I need is a narrow-band filter, so with high-Q. However, I am having difficulty calculating loaded-Q. Found this definition:</p> <blockquote> <p>Loaded Q (Working Q): A term that defines the percentage of the 3db bandwidth of a Bandpass Filter. Q=Center Frequency (Fc) in Hertz/ 3dB Bandwidth in Hertz</p> </blockquote> <p>Now, I believe that Fc in my case is 433.92x10^6 Hz (right?), but not clear on what "3dB Bandwidth in Hertz" means.</p> <p>However, I've read other posts which talk about the difficulties of working with any circuit at such high frequencies (s.a. in the UHF range), due to parasitic effects, and if I understood it correctly, then the extensive use of ground-plane, i.e. definitely must work with at least 2-sided PCB, and nothing that can be done on a breadboard.</p> <p>While I cannot claim to have done everything possible to answer the question myself, but I trust that I've put some reasonable effort in going as far as my current level of knowledge would allow. I understand that one expected answer is, this is clearly beyond your league, and you shouldn't even attempt such a thing. Fair enough, but could I insist ? Maybe in a different forum, or say over chat ?</p>
Noob asking for circuit design help - any right way?
<p>I myself saw it as more RF/MAC layer oriented, not windows configuration, so it's an engineering question.</p>
2527
2012-12-20T00:34:12.963
|discussion|
<p>This question just showed up:</p> <p><a href="https://electronics.stackexchange.com/q/51774/2028">Does Wifi signal strength affect link speed?</a></p> <p>It's one of those borderline questions that at first glance make you want to vote to close. Then I thought, well, maybe it should be migrated to SuperUser (it's about Windows' interaction with wireless networking)...</p> <p>But maybe it's more about general RF and antenna orientation, which might be better migrated to Physics. But I don't think Physics is quite as appropriate for radio spectrum questions when it clearly involves some sort of electronic circuit.</p> <p>At the end of the day, I feel it's more about a consumer electronics application, not necessarily an EE challenge or problem.</p> <p>Should it be addressed at all, or closed as off topic?</p>
Radio/antenna questions; bordering on off-topic
<p><code>&lt;sub&gt;</code> and <code>&lt;sup&gt;</code> tags would do the job too</p> <p>I<sup>2</sup>R<sub>load</sub> is produced by <code>I&lt;sup&gt;2&lt;/sup&gt;R&lt;sub&gt;load&lt;/sub&gt;</code></p>
2544
2013-01-07T09:40:12.560
|support|
<p>In my answer <a href="https://electronics.stackexchange.com/a/53212/9730">here</a> I have tried to get P = I^2 * R to show the 2 in super script, several different ways, like P = I^{2} * R as mentioned <a href="https://electronics.meta.stackexchange.com/a/438/9730">here</a>.</p> <p>How is it done?</p>
I can't seem to get superscript to show up, like I squared should be I^2
<p>This indeed is possible with LaTeX using the <a href="http://www.ctan.org/pkg/tikz-timing" rel="nofollow"><code>tikz-timing</code></a> package. It isn't included in the markdown editing though. But this might be possible..? It would be really useful, the <code>tikz-timing</code> is used by every professional organization to draw logic diagrams in datasheets, application notes, etc...</p>
2547
2013-01-09T08:48:32.040
|discussion|support|
<p>In EE all we need to draw some waveforms to ask or answer questions.</p> <p>Is this could be done with TeX? Or what are the software's you people are using?</p> <p>Could we have add some tools to draw waveforms in here?</p>
Drawing Waveforms
<p>If you are wanting to thank someone just up vote. Comments such as, "Thanks for the help." are just noise. If it is helpful, up vote. If it solves your problem, accept. If you need further information or have input to give them about the situation please post a comment explaining.</p> <p>Although I am sure someone will say that saying thanks is appreciated, that is a duplicate of up voting with less lasting value. It is the same as signing a signature, don't do it because the site puts a little signature on your post for you.</p>
2551
2013-01-10T01:20:45.553
|discussion|
<p>Do I need to try to pay attention to comments like thanks with a sentence or something afterward.</p> <p>Sorry for a stupid-ish question, but I'm very new with regards to social media sites.</p>
Does leaving thanks type comments get automatically/mod deleted at some point or should I be deleting them myself after?
<p>Even though it is short circuited, it looks nice. The display just needs a blue tint for a VFD look.</p>
2557
2013-01-11T15:01:10.510
|discussion|
<p>I realized it's a breadboard, and the ICs are not across the channel in the middle. It's been bothering me since.</p>
Does it bother anyone else that the site banner shorts pairs of IC pins together?
<p>It is the number of votes to close <em>you</em> have remaining. It is similar to the maximum Q&amp;A votes and flags in that it resets every 24 hours. Right now, with our site size and volume, you would only exhaust those close votes if you went through a full review queue... it shouldn't be an issue.</p> <p><a href="https://electronics.stackexchange.com/faq#close">https://electronics.stackexchange.com/faq#close</a></p>
2573
2013-01-15T23:44:11.293
|discussion|
<p>When I bring up the dialog to close a question, at the bottom it says "24 votes remaining to close". Obviously this doesn't refer to the number of votes required to close the question since only five are needed.</p> <p>Is this the number of votes <em>I</em> have available to close questions? If so, how or when does the number get reset?</p> <p>I looked in the FAQ for both the site and here on meta, and couldn't find anything about this.</p>
What does "24 votes remaining to close" mean?
<p>A tag for a single chip is usually too granular for the tag system. A lot of tags get created because the threshold for adding one is fairly low, but that doesn't mean they need to stick around. I went ahead and removed the tag from the only question that used it, and the tag will eventually be automatically cleaned up. This particular tag is definitely too granular since there are four chips in the family that are very closely related - a tag on only one would exclude the other three. (18F4550, 18F4455, 18F2455, 18F2550)</p> <p>You already can search for specific parts (try searching for 18f2550 on the main site - it still works without any questions being tagged). The tags for a question shouldn't be the place where you get important information regarding the question (e.g. what chip they are using) - that should already be in the question. </p>
2587
2013-01-20T12:43:15.130
|discussion|tags|tag-cleanup|
<p>There seems to be a <a href="https://electronics.stackexchange.com/tags/pic18f2550/info">pic18f2550 tag</a>. Why is that? It isn't really used. Shouldn't it be merged with the <a href="https://electronics.stackexchange.com/tags/pic/info">pic tag</a>, before we have a tag for every PIC?</p>
PIC18F2550 tag merging with PIC?
<p>Most of the time it's not a big enough deal to do anything about. This case is as good a example as any. Yeah, the comment is mostly pointless, but it's not rude, offensive, or really badly written to the point I'd go out of my way to get rid of it.</p> <p>Let's keep this in perspective. Dave's well written, complete, and straight forward answer still stands, and anyone looking at this Q+A later will read that first. So it's a little more disk space, screen space, and noise at the <i>bottom</i>, but that's about it.</p> <p>I've seen far worse comments. Sometimes people seem to chime in only because someone else wrote the answer they would have before they got there. That might be what happened here. Sometimes, someone is having a bad day and just wants to be <a href="https://electronics.stackexchange.com/a/42788/4512">contrary</a>. That exchange actually started with the first comment saying the post was "staggeringly wrong", which was later edited to "I disagree". When you get a bunch of rep, your posts seem to attract wannabes that look for any excuse to find something wrong so that they can look smart. The technical vigalance is good, but the resulting comment chain is often messy noise.</p> <p>So on a scale of 0-10, I'd say this one ranks about &frac12;.</p> <p><a href="https://electronics.stackexchange.com/a/55879/4512">Here</a> is another example of a completely pointless comment. We really don't care what this guy "read somewhere", and otherwise seems to be just <i>"Looky me world!"</i> without any relevant information. However, it's not worth doing anything about. I could reply with <i>"... and your point is?"</i> but that would be adding more noise.</p>
2592
2013-01-22T17:40:55.920
|discussion|comments|
<p>I'm looking at for example the second comment on <a href="https://electronics.stackexchange.com/questions/55767/who-receives-the-value-returned-by-main#55773">Who receives the value returned by main()?</a>. The commenter simply says the same as Dave in the answer, with the only addition that C wasn't meant for OS-less microcontrollers - but that is of no importance to the answer and also follows logically from the answer.</p> <p>What should we do with these comments? Flagging and removing is rude(?), but on the other hand, I don't see they're a useful addition.</p>
What should we do with comments that just rephrase the post?
<p>Keep in mind that the primary interface to StackExchange questions is Google. Given that you've accepted Dave Tweed's answer, if I were to Google search for the original question you asked and the result was not helpful to me, there is a problem. Rephrase the question such that the accepted answer answers the question of someone finding the question via Google.</p>
2595
2013-01-23T16:19:42.863
|discussion|
<p>I asked <a href="https://electronics.stackexchange.com/questions/55871/how-do-i-design-a-triac-rms-power-limitter#comment105572_55871">How do I design a triac rms power limiter?</a>. <strong>Dave Tweed</strong> answered, in a comment, something that solved my problem. What now? Should I reform the question to make the answer a direct answer to the question. Should I ask Dave to repost his thoughts as an answer, so that I can accept it? </p>
I asked the wrong question and got the right answer. What now?
<p>You no longer need to add <code>\cancel</code> anc <code>\bcancel</code> yourself, as they are now part of the <code>cancel</code> extension that was made available as part of MathJax v2.0. If you are running your own site, you can include this in your TeX extensions:</p> <pre><code>MathJax.Hub.Config({ TeX: {extensions: ["cancel.js"]} }); </code></pre> <p>Otherwise, if you are using an SE site with MathJax that doesn't load this for you, use</p> <pre><code>\require{cancel} </code></pre> <p>within your mathematics and that will load it for you. E.g.</p> <pre><code>$$\require{cancel} \cancel{2-2}$$ </code></pre> <p>gives</p> <p>$$\require{cancel} \cancel{2-2}.$$</p>
2599
2013-01-23T18:36:56.627
|feature-request|mathjax|
<p>Some MathML / MathJax sites <a href="https://groups.google.com/forum/?fromgroups=#!topic/mathjax-users/tuFHn3EkSic" rel="nofollow">have support for strikethrough</a> enabled:</p> <pre><code> MathJax.Hub.Register.StartupHook("TeX Jax Ready",function () { var TEX = MathJax.InputJax.TeX; var MML = MathJax.ElementJax.mml; TEX.Definitions.macros.cancel = ["myCancel",MML.NOTATION.UPDIAGONALSTRIKE]; TEX.Definitions.macros.bcancel = ["myCancel",MML.NOTATION.DOWNDIAGONALSTRIKE]; TEX.Parse.Augment({ myCancel: function (name,notation) { var mml = this.ParseArg(name); this.Push(MML.menclose(mml).With({notation:notation})); } }); }); </code></pre> <p>This code enables the use of \cancel and \bcancel respectively. These macros don't seem to work on our site:</p> <p>\$ \cancel{(2 + 2)} \$</p> <p>Since some of our answers deal with derivations, having the ability to do a strikethrough may help with translating written derivations into electronic format.</p> <p>There is a \not function, but it strikes through only a single character (not an expression). </p> <p>Thoughts / comments?</p>
MathML support for strikethrough
<p>You should definitely ask and answer your own questions. <a href="http://blog.stackoverflow.com/2011/07/its-ok-to-ask-and-answer-your-own-questions/">This is explicitly encouraged.</a></p>
2605
2013-01-26T06:48:22.450
|discussion|
<p>I've noticed recently that <a href="https://electronics.stackexchange.com/users/17608/phil-frost">Phil Frost</a> has posted a few questions and immediately answered no doubt as a form of tutorial, an example is <a href="https://electronics.stackexchange.com/questions/56170/how-can-suddenly-stopping-a-spinning-motor-cause-my-supply-voltage-to-shoot-up-w">How can suddenly stopping a spinning motor cause my supply voltage to shoot up with the back-EMF can't exceed the supply voltage?</a></p> <p>I think they are excellent, but I guess strictly interpreted based on the FAQ statements such as "You should only ask practical, answerable questions based on actual problems that you face" maybe they aren't real questions considering he already knew the answer? Please note that I am not in any way saying they shouldn't be allowed, but maybe the FAQ or elsewhere should clarify it a bit.</p> <p>The reason I ask is that a few times in the past I've had problems I've been considering asking, but after further research have solved myself at which point I don't consider them a question but the solution I found may be interesting to the general community. So I'm wondering the guidelines for when it is or isn't OK to post them in that format?</p>
Self-answering questions as a form of tutorial
<p>I say EE is the place! Since the hardest (or at least the most complex part) of encrypting/decrypting a PIC's memory is in the implementation/access (as compared to encrypting some data in an asp.net environment.)</p> <p>Plus, this site doesn't get enough cool questions! :-) and I'm not a member of cryptography...</p>
2612
2013-01-29T07:10:38.013
|discussion|on-topic|
<p>Let's say I have a PIC and an EEPROM memory IC. I store sensitive data in the EEPROM. If I want to encrypt the data, should I post it on EE or on <a href="https://crypto.stackexchange.com/">Cryptography</a> as a request for a fast and low memory cost encryption?</p>
Are data encryption ideas for PICs on-topic?
<p>The initial revision of your question had one reference to "Linux" and zero references to any sort of hardware. Even though you knew you were wondering about real-time scheduling for an FPGA project, that wasn't obvious to the people reading the question. Once you added that information, you got seven upvotes (more than most questions) because it was obvious how this question related to electrical engineering.</p>
2614
2013-01-30T04:57:14.277
|discussion|close-reasons|specific-question|
<p>I want to learn real-time programming. I asked this and it's near getting closed:</p> <p><a href="https://electronics.stackexchange.com/questions/56491/what-features-distinguishes-a-real-time-core-from-other-types-of-os">What features distinguishes real-time from other types of os?</a></p> <p>Should I not at all have asked this question or could it have been in scope at some other Q&amp;A site?</p> <p>Why all the close votes? Becuase its mostly software I'm asking about?</p> <p>But I got real good answers though, that's why I asked on EE.</p> <p>Thank you</p>
Was my real-time programmering question off-topic?
<p>I think it's important to note that C for EE can be for x86, but is many times embedded. So there is a larger emphasis on optimization, memory conservation, low level access, etc with which many of the C experts on SO are not familiar.</p>
2622
2013-02-02T08:04:02.597
|discussion|on-topic|
<p>I asked <a href="https://electronics.stackexchange.com/questions/56058/c-coding-design-function-pointers">C coding design - function pointers?</a> on which EJP commented it's off-topic and/because it belongs on SO. Kortuk says no:</p> <blockquote> <p>@EJP I disagree. Just because there is an overlap does not mean it has to be on one site or the other. It is asking questions related to the design and programming of low level embedded systems, that seems on topic either place.</p> </blockquote> <p>Now the same thing happens at this answer: <a href="https://electronics.stackexchange.com/questions/56858/absolute-address-of-a-function-in-microchip-xc16#answer-56861">Absolute address of a function in Microchip XC16</a></p> <p>Who's right, EJP or Kortuk?</p>
Are pure C questions on-topic?
<p>We heavily support arduino on site, previously this has resulted in the arduino site being closed as a duplicate. </p> <p>I see no issue in the more detailed tag wiki linking to a proposed site, but I would not guarantee you are getting a site either.</p>
2623
2013-02-02T16:23:14.137
|discussion|
<p>The <a href="http://area51.stackexchange.com/proposals/49538/arduino">Arduino site proposal</a> would interest a lot of the users of this site, so is it alright to promote it through the relevant tag wiki(s) ?</p>
Is it alright to promote a site proposal through the relevant Tag wikis?
<p>I didn't downvote that question, and I don't see a obvious reason to downvote it. You left some junk "NB>" characters in there, but that relatively small editing oversight doesn't rise to the level of deserving a downvote in my opinion. I upvoted it to compensate for one of the downvotes.</p> <p>To answer your question, there is no real incentive to downvote other than the desire to keep the site clean. We get a lot of sloppily worded questions with baby talk instead of real English words, and I downvote those regularly, for example. You don't gain or loose any rep by voting on questions. I guess there is a very weak incentive in that votes on questions count towards your total votes cast, and I think there is a badge or two when you get to certain levels. However, badges don't mean much here and I doubt anyone is going around randomly voting to gain a badge.</p> <p>Sometimes downvotes happen for mysterious reasons. I still think this site would be better off if all votes were public, but those that run this site disagree. I don't have a problem with someone politely asking why they were downvoted, and I have done that myself (usually don't get any answer), but calling the downvoters children without knowing their reasons is just as wrong. That in itself can be deserving of a downvote in my opinion.</p>
2630
2013-02-04T13:46:32.920
|discussion|moderation|voting|down-votes|
<p>Here is a <a href="https://electronics.stackexchange.com/questions/57012/rfid-coil-ends-broke-can-i-resolder">question</a> which I asked earlier today, had got a reasonably good answer and was waiting until at least 24 hours as I usually do, to give time for any other comments/feedback, when I suddenly see a series of 2 down-votes, within a short span of time. No comment left behind. Clearly, the downvote wouldn't serve the purpose i.e. neither does it improve the question, nor the asker's way of position questions for future.</p> <p>It makes me think of only 2 possible reasons for such downvotes:</p> <ol> <li>Down voter had an incentive to downvote.</li> <li>Down voter dislikes me ! Yes, I've met many children in grown-up bodies.</li> </ol>
Are their incentives for downvoting questions?
<p>This is a good example of needing to make sure everyone understands your terms. The only "Redstone" I know is a rocket [<a href="http://en.wikipedia.org/wiki/PGM-11_Redstone" rel="nofollow">PGM-11</a>], and I've never heard of "Minecraft". Simply expecting people to know what those are is going to sound arrogant and get your question downvoted and closed fast.</p> <p>The solution (if this is appropriate to this forum at all) is to make sure you define your context. Actually that's always the case, just that you can assume some domain-specific context here. For example, you don't need to define what a resistor or capacitor is here. If your question relies on people understanding whatever Redstone and Minecraft are, you'd better give a short introduction. And no, just a link won't do it. I'm not going to follow a link for what I think should be basic mandatory information in your question. If you give a quick introduction so we know basically what it's about, then you can follow with a link to details.</p> <p>You want the thought process of someone who doesn't know these things to be <i>"Oh, that's what he's talking about. I don't know anything about this so I'll just go on to the next question"</i>, as apposed to <i>"What a arrogant prick! He thinks his obscure problem domain is so important that I'm supposed to know all about it or research it on my own? Not gonna happen. Downvoting and voting to close as off topic or not a question."</i>.</p> <p>The more off center of electrical engineering you get, the more you have to make sure you define whatever jargon you use. Of course if you need to do a lot of defining, then it's a good clue the whole question may be off topic.</p>
2633
2013-02-05T01:32:32.003
|discussion|asking-questions|
<p>The title kind of stands for itself. I mean, you can make some <a href="http://www.youtube.com/watch?v=wgJfVRhotlQ" rel="nofollow">pretty advanced things</a> with redstone in Minecraft. When it comes to complicated stuff like making computers with redstone, would questions related to that be appropriate for EE?</p>
Are questions about redstone creations in Minecraft appropriate for EE?
<p>Very few fonts give an unambiguous distinction between 'o', 'O', and '0'. The only one that I've seen is basically a monospaced font, the same one seen in the edit window and in <code>c0de</code> blocks.</p> <p>I believe that this is a communication issue and not so much a font issue. We have been dealing with how to clearly and accurately convey engineering terms and quantities for a long time, and there are a few guidelines to ensure that people correctly interpret what was written. These have been reasonably standardized by the primary organization for EE's, the IEEE. This <a href="https://development.standards.ieee.org/myproject/Public/mytools/draft/styleman.pdf" rel="nofollow">style guide</a> gives guidelines for clear and unambiguous writing so that you can be understood regardless of the font choice. </p> <p>To pull out and summarize a few relevant points from section 13.2:</p> <ul> <li>Separate the quantity from the units with a space. ( 73 ms )</li> <li>Add a leading zero when the magnitude is less than one. ( 0.25 ms )</li> <li>Ranges should never have a dash because it may be confused with subtraction. ( 0 ms to 10 ms )</li> </ul>
2635
2013-02-05T06:25:17.067
|feature-request|design|formatting|
<p>In a comment, I wrote: </p> <blockquote> <p>Absolutely 0ms isn't possible with any circuit.</p> </blockquote> <p>Here, in a question, you can see a slight difference between the zero and the o: 0o. But in a comment you can't: <a href="https://electronics.stackexchange.com/questions/57025/0-10-second-on-time-with-a-potentiometer/57027#comment108427_57027">0-10 second on-time with a potentiometer</a></p> <p>Can we please have a font that shows the difference between o and 0 in both questions/answers and comments? It's very important for our users, I believe.</p> <p>I've read <a href="https://electronics.meta.stackexchange.com/questions/481/new-design-launched/513#513">this old answer</a> on meta that states the same. You can see there that this can be workarounded with either \$\LaTeX\$ or <code>code</code> digits, but I'd say that's overkill for simple things like "0ms". </p>
The zero is an o in comments
<p>Yes.</p> <p><img src="https://i.stack.imgur.com/GeSHI.png" alt="enter image description here"></p>
2638
2013-02-05T18:17:06.723
|discussion|feature-request|
<p>We got a little discussion at this post: <a href="https://electronics.stackexchange.com/questions/57184/amplifier-vpp-calculation">Amplifier Vpp calculation</a>. The Photon says yes/no questions aren't good for SE and I agree. They aren't right? (1st question)</p> <p>If you agree, where is it in the FAQ? I can't find it. (2nd question)</p> <p>If it isn't there, should it be added? (3rd question)</p>
Yes / no questions
<p>I think that certain tags should be created as mentioned in the question. The tags could be :</p> <ul> <li><p><code>main-faq</code> or <code>site-faq</code> : Questions about the <a href="https://electronics.stackexchange.com/faq">main FAQ</a></p></li> <li><p><code>meta-faq</code> : Questions about <a href="/questions/tagged/faq" class="post-tag moderator-tag" title="show questions tagged &#39;faq&#39;" rel="tag">faq</a> </p></li> <li><p><code>faq</code> : The actual <a href="/questions/tagged/faq" class="post-tag moderator-tag" title="show questions tagged &#39;faq&#39;" rel="tag">faq</a> questions on Meta</p></li> </ul> <p>These are similar to tags used on <a href="https://meta.stackoverflow.com/tags">MSO</a></p>
2639
2013-02-05T18:18:49.870
|support|feature-request|
<p>It appears to be that the <a href="https://electronics.meta.stackexchange.com/questions/tagged/faq">FAQ tag</a> on meta is used for the questions that are most frequently asked on the meta site. What if I want to post something about the <a href="https://electronics.stackexchange.com/faq">FAQ page</a> on the main site, like <a href="https://electronics.meta.stackexchange.com/questions/2638/yes-no-questions">Yes / no questions</a>? Is that possible? If it isn't, can it be added?</p> <p>So can there be a new tag <code>main-faq</code> or <code>about-the-faq</code>?</p>
FAQ tag not about main site FAQ
<p>Battery life calculations are definitely part of electrical engineering, so your question is clearly on topic.</p> <p>However, this particular question needs work. Clean up the English, use real sentences, and keep in mind we don't have the same context about your project that you do. Your second line isn't a sentence and is particularly confusing. I might downvote this if you post it as is. It's not clear how the battery you mention relates to the "device". You haven't even said what voltage the battery is.</p> <p>This could be closed as not a real question in its current form.</p>
2645
2013-02-06T02:01:08.090
|discussion|
<p>I have the following question:</p> <pre> The rechargeable battery info: 1900 mAH A device that have input of 3V DC and output of 5 V / 500mA If two rechargeable battery is insert into the device, I can expect that it can provide me ( 1900 mAH / 500 mA of 3.8 hours ) of output. So, if the device only provide me 1 hour of output, can I say that the circuit is faulty (suggestive that the device provide 1900 mA of output)? </pre> <p>So, should the above question be posted in Electrical Engineering or other place in Stackexchange?</p>
Is question on battery discharge can be asked in here?
<p>We each have different opinions and thresholds for what we think is appropriate or not. That's in part why there is a voting system for the more drastic actions. In that sense, as long as you do what you believe to be correct, judgement calls can't technically be "wrong".</p> <p>That said, I disagree with your judgement in this particular case. While the answer doesn't strictly speaking answer the question, and therefore you can make a argument for it being "not an answer", I wouldn't have flagged it as such. I personally don't have a problem with answers that don't directly address the question as long as they provide some relevant information. OPs often don't ask what they really want to know, and often don't know that they don't know something important. I personally consider answers that provide such information as being relevant, useful to the OP and the site, and not deserving of being removed.</p> <p>There is more discussion of this issue <a href="https://electronics.meta.stackexchange.com/q/1428/4512">here</a>.</p>
2648
2013-02-07T14:10:49.533
|support|
<p>I reviewed ADI_iCoupler answer on <a href="https://electronics.stackexchange.com/questions/57302/digital-isolation-comparison">Digital Isolation comparison</a>. I consider this answer to be <em>helpful</em>, but it doesn't answer the question. I marked it as not an answer, but don't know whether that was the right thing to do. Apart from this example, what should you do as a reviewer with helpful answers that aren't really answers?</p> <p><img src="https://i.stack.imgur.com/FP6YJ.png" alt="Post being discussed"></p>
Useful no-answer answers
<p>I would suggest not an answer, I dont have a strong preference and your flag will be marked as valid regardless which you pick.</p>
2654
2013-02-08T18:19:40.370
|support|
<p>It's clear that <a href="https://electronics.stackexchange.com/questions/22944/can-a-110v-spa-pump-curcuit-control-a-220v-spa-heater-that-is-on-its-own-220v-c#comment109143_57471">answers in the wrong language</a> have to be flagged. But should they get flagged as not an answer, very low quality, spam, not welcome or other?</p> <p>I marked this one as other with the note "Wrong language". </p>
Wrong language flag
<p><img src="https://i.stack.imgur.com/o8aZ5.png" alt="XKCD comic"></p> <p>(<a href="http://xkcd.com/435/" rel="nofollow noreferrer">from XKCD</a>)</p> <p>The analogy to Arduino and Electrical Engineering should be clear.</p> <p><a href="https://electronics.stackexchange.com/questions/57824/how-do-i-get-5v-for-loud-noise-0v-for-silence-from-electret-microphone-or-oth">This question asked by an Arduino user</a> got me thinking about a different concern. The OP there is making a very good effort to break out of the Arduino abstractions and move to the right on this chart, to electrical engineering.</p> <p>The difficulty, it seems, is that there's a clear incremental path to move left on this scale, but to move to the right is more difficult. Sure, there's a huge body of knowledge that a physicist possesses that a mathematician does not, but an experienced mathematician wanting to move to the left will start at basic physics, while a physicist trying to move to the right will start at advanced mathematics.</p> <p>The analogy breaks down here, because you'd expect every physicist understands that their field is built upon mathematics, and they'd know well enough that they have to jump <em>all the way back</em> to basic mathematics and work to the left to bridge the gap. However, I've had some Arduino users tell me they know some electronics, as if knowing to use an Arduino is like basic electrical engineering. But it's not: knowing the <em>fundamentals</em> of a field is not the same as knowing the <em>applications</em> of a field. Consequently, it seems to take a bit of tough love to get people wanting to move from Arduino to electrical engineering to realize they lack the fundamental understanding necessary to implement a circuit that doesn't come ready-made on a shield, and that they need to back off and focus on the fundamentals.</p> <p>It would seem to me that having a separate site for Arduino would benefit both communities by making that movement more clear.</p> <p>I'd be interested to know if this concept of jumping backwards being harder resonates with anyone else. I learned assembly first and Python last, which isn't how most people do it, but I tried going in the other direction of abstraction and found it really hard. Am I just an odd learner?</p>
2656
2013-02-10T03:20:08.670
|discussion|
<p>The <a href="http://area51.stackexchange.com/proposals/49538/arduino">Arduino site proposal</a> is moving forward at a good rate. As a part of the scope of the Arduino site will cover questions that currently fall in the scope of EE, what will be the benefits (and problems) in having a separate Arduino site ?</p> <p><a href="https://area51.meta.stackexchange.com/a/9140/70968">This post</a> discusses this issue in some detail, and I mostly agree with the views given there.</p> <hr /> <h2>Updates</h2> <ul> <li>March 28 - The proposal is now in the commitment phase.</li> <li>April 9 - The proposal is now in private beta.</li> </ul>
What will be the benefits in having a separate Arduino site?
<p>I would like to see more salvage questions but some sort of guideline needs to be set up. After reading this thread I posted a topic about salvaging LCDs and it was closed within a day. They cited it as "too localized" which was quite puzzling considering that I never mentioned where I am and devices which parts can be salvaged from can easily be shipped all over the world from sites like eBay. The short description of what "too localized" mean was equally unhelpful and the included link to the FAQ only repeated this verbatim without any further elaboration so it was worthless as well. At the moment I'm puzzled as to what would constitute a valid salvage question.</p>
2658
2013-02-10T21:33:09.603
|discussion|
<p>I asked <a href="https://electronics.stackexchange.com/q/57625/2028">a question about salvaging parts from used/junk equipment</a>, and it was negatively received.</p> <p>Many hobbyists cannot afford to buy new parts (such as pneumatic actuators, solenoids and stepper motors) so they often salvage them from equipment. I know I've desoldered plenty of relays and capacitors for re-use.</p> <p>I realize <strike>all</strike> <strike>most</strike> <strike>many</strike> some shopping questions are off-topic, but is advice on how to acquire parts from used equipment similarly off-topic?</p> <p><strong>Edit:</strong></p> <p>Quantified the comment about shopping questions a bit better. :)</p>
Salvage questions. On topic?
<p>Yes, click the share button below the answer/question for a link.</p> <p><img src="https://i.stack.imgur.com/NY9gx.png" alt="Image of the button push"></p> <p>It includes your userid in the link so that it can track and give badges for how many you refer to the question/answer. It is the last number, you can remove it manually if you like.</p>
2669
2013-02-12T14:37:32.770
|support|
<p>Sometimes (on meta as well as on main) I want to link to an answer. It is possible to get a link to a comment: by clicking on the timestamp you get a link with on the end something like <code>#comment109723_57751</code>. Do we have something similar with answers? </p>
Links to answers
<p>In response to <a href="https://meta.stackoverflow.com/questions/303812/discourage-screenshots-of-code-and-or-errors">this Meta Stack Overflow request</a> we have implemented a change that will <em>sort of</em> allow low rep users to post images. </p> <p>If a user with &lt;= 15 rep tries to post a screenshot, they are presented with a message in the image uploader:</p> <blockquote> <p><a href="https://i.stack.imgur.com/LqY9v.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/LqY9v.png" alt="enter image description here"></a></p> </blockquote> <p>If they are on a site where image embedding is blocked for new users, we also implemented a change that if a user has less than the <a href="https://stackoverflow.com/help/privileges/new-user">new user privilege</a> and they add an image, we will now include a link to the image and a message.:</p> <blockquote> <p><a href="https://i.stack.imgur.com/2236z.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/2236z.png" alt="enter image description here"></a></p> </blockquote> <p>Instead of embedding the image, the post will now have a link. Previously these users would get an error message when they attempted to post. </p>
2671
2013-02-12T14:44:31.097
|feature-request|status-completed|
<p>New users can't post images. On websites like StackOverflow this isn't really necessary, and thus a good way to prevent new users from spamming unwanted images. However, on EE.SE, we often need a schematic, a pinout diagram, an image of a pcb or circuit. Isn't it a good idea to remove the 10 reputation restriction for images? Or perhaps a limit on the amount of images for new users?</p> <p>Related: <a href="https://electronics.meta.stackexchange.com/a/2632/17592">https://electronics.meta.stackexchange.com/a/2632/17592</a> </p>
Images for new users
<p>These notification messages should now disappear on their own without any clicking required at all.</p>
2675
2013-02-15T10:15:30.103
|feature-request|bug|status-completed|
<p>A few days ago, we got yellow messages like "please consider adding a comment if you think this post can be improved". When you clicked on such a message, it disappeared. With the new style (red), this isn't possible anymore. Some messages, like "You may only submit a comment vote every 5 seconds." have a cross on the right to dismiss it. </p> <p>I got used to dismissing by clicking and found it way easier than the cross we now have. Is there a good reason for removing the click-to-dismiss feature? If not, can that feature please return? Otherwise: Why haven't all the messages a cross? Can they all have one if there is no good reason?</p>
Dismissing messages
<p>Well, This one is fairly simple. I went ahead and took care of it, but it's something practically everyone can do. </p> <p>Someone misspelled the tag (not clicking on the suggested existing <a href="https://electronics.stackexchange.com/questions/tagged/comparator" class="post-tag" title="show questions tagged &#39;comparator&#39;" rel="tag">comparator</a> tag). So edit the tags on the one question that has that tag, and change it from <a href="https://electronics.stackexchange.com/questions/tagged/comparitor" class="post-tag" title="show questions tagged &#39;comparitor&#39;" rel="tag">comparitor</a> to <a href="https://electronics.stackexchange.com/questions/tagged/comparator" class="post-tag" title="show questions tagged &#39;comparator&#39;" rel="tag">comparator</a>. The system will automatically purge tags with zero questions in due time. If it's something more serious than 2-3 tags, then that would be a good time to bring it to Meta or notify a moderator.</p>
2678
2013-02-16T02:23:55.897
|support|
<p>Tag comparitor needs to be fixed. Somehow whoever created misspelled the word.</p> <p><a href="https://electronics.stackexchange.com/questions/tagged/comparitor">https://electronics.stackexchange.com/questions/tagged/comparitor</a></p> <p>How to fix or report such fixing requests?</p>
How to fix or report such fixing requests?
<p>You can just re-click the upvote mark to undo your vote. However, after a certain time (how long, I do not know) your vote is "locked": you cannot undo it (or change it into a downvote) until the answer is edited.</p>
2680
2013-02-16T12:03:58.480
|support|
<p>I recently up-voted an answer but upon further reflection while not a really bad answer it didn't address a few aspects of the question. When I pressed down-vote it went from +1 to -1 instead of going back to zero.</p> <p>I didn't think the answer deserved -1 so up-voted again to +1 but wondered if there is a way to just undo a vote so it would have gone back to zero in the above case? Not sure if I've missed something or if anyone else thinks it may be worthwhile feauture?</p>
Cancelling an answer up-vote
<p>Quoting the Stack Overflow <a href="https://meta.stackexchange.com/questions/67397/list-of-all-badges-with-full-descriptions">FAQ</a> for badges :</p> <blockquote> <p>Earned at least 200 reputation in a single day. All reputation activities, including up and down votes, accepted answers, bounties, and suggested edits count towards this badge except for association bonuses, which do not count.</p> </blockquote> <p>Which is mentioned in this <a href="https://meta.stackexchange.com/a/68258/200868">answer</a>.</p>
2682
2013-02-16T13:39:53.920
|support|
<p>As you can see on <a href="https://electronics.stackexchange.com/users/17592/camil-staps?tab=reputation">my reputation history</a>, on January 8, I got 266 reputation. I believe I should get the <a href="https://electronics.stackexchange.com/badges/25/mortarboard">mortarboard badge</a> for that, but <a href="https://electronics.stackexchange.com/users/17592/camil-staps?tab=badges">I don't have it</a>. </p> <p>Am I misunderstanding something?</p>
Why haven't I the mortarboard badge?
<p>See <a href="https://electronics.meta.stackexchange.com/questions/2686/reviewing-and-flagging-answers-causes-the-forum-software-to-automatically-add-a/2689#2689">Reviewing and flagging answers causes the forum software to automatically add a comment impersonating me?</a></p> <p>The "recommend deletion" dialog is pretty clear about the fact that it will leave a comment on your behalf should you ask it to do so for you.</p> <p><img src="https://i.stack.imgur.com/Y9KCM.png" alt="enter link description here"></p>
2688
2013-02-17T07:09:12.457
|feature-request|
<p>Basically, The feature that when you flag a post, it also posts a comment to said post detailing why you flagged it, has recently been added to the SE software.</p> <p>This caught me a bit by surprise, as it changes the existing behaviour of the SE site, and I wound up having people respond to comments that I wasn't aware I posted.</p> <p>I think it would be nice if there was some information somewhere in the flagging dialog that flagging the post will auto-add this comment. </p> <hr> <p><sup><sub>For that matter, can we get an explanation of what specific actions trigger this macro in the first place? Is it for every possible post flag, or only some of them?</sub></sup></p>
Let the user know that flagging a Question/Answer will cause you to automatically add a comment
<p>I don't think it's necessary, because it is a meta tag. Using the points made in the link the OP posted:</p> <blockquote> <p>If the tag can’t work as the only tag on a question, it’s probably a meta-tag. Every tag you use should be able to work, more or less, as the only tag on a question.</p> </blockquote> <p>It wouldn't work as an only tag, since it would be describing what the question is not, an infinite possibility of things.</p> <blockquote> <p>If the tag commonly means different things to different people, it’s probably a meta-tag. </p> </blockquote> <p>How is the poster to determine if the tag is to be used? </p> <ul> <li>Every non-arduino question?</li> <li>Perhaps they would prefer to not use an arduino</li> <li>Perhaps they wish to explicity exclude arduino </li> </ul> <p>Which one is right?</p> <p>These two tests confirm that not-arduino is in fact a meta tag, and the article mentions that meta-tags are explicitly discouraged. </p> <p>Additional points:</p> <ul> <li><p>The arduino tag (or lack therof) should be adequate enough.If the question does not require the use of an arduino, the poster could just not include the tag. </p></li> <li><p>The tag sets an example for the creation of other "not-tags".</p></li> <li><p>The search function should include the ability to specifically exclude tags (if it does not do so already), so that a searcher may exclude the arduino tag from any post results if they are specifically looking for non-arduino posts.</p></li> </ul>
2691
2013-02-17T13:44:21.617
|discussion|status-completed|tags|
<p>Today, the <a href="https://electronics.stackexchange.com/questions/tagged/not-arduino" class="post-tag" title="show questions tagged &#39;not-arduino&#39;" rel="tag">not-arduino</a> tag was added to a bunch of questions. Apparently it is for questions which do not want an Arduino solution.</p> <p>Is it really necessary? If someone doesn't want to answer arduino questions, they can just ignore the <a href="https://electronics.stackexchange.com/questions/tagged/arduino" class="post-tag" title="show questions tagged &#39;arduino&#39;" rel="tag">arduino</a> tag (the 'ignore' functionality makes the posts disappear).</p> <p>Besides, it seems to me like <a href="https://electronics.stackexchange.com/questions/tagged/not-arduino" class="post-tag" title="show questions tagged &#39;not-arduino&#39;" rel="tag">not-arduino</a> is a <a href="http://blog.stackoverflow.com/2010/08/the-death-of-meta-tags/">meta-tag</a>:</p> <blockquote> <p>The reason meta-tags are a problem is that they do not describe the content of the question. They describe some other aspect of the question, like the author’s skill level, or the author’s motivation for asking it, or generally what “kind” of question it is (poll, how-to, etc.).</p> <p>Meta-tags are actually a subset of a larger problem that I usually call dependent tags. These are tags that don’t say anything by themselves – you can’t tell what the question is about unless they’re paired with some other tag (or several of them). </p> </blockquote> <p>It seems like a dependent tag to me. It is also pretty similar to "author's skill level", in the sense that it is filtering out answers based on what the author will and won't accept; whereas other people/visitors may (after all, we write answers for visitors to read, not just to help the OP).</p>
Is [not-arduino] necessary? Is it a meta tag?
<p>As Jeremy said in the comments, these error messages still have Xs to dismiss; but they also can be closed by clicking them (as before), and now fade away after a set period of time.</p> <p>Think some of the colors may still be being tweaked though.</p>
2696
2013-02-19T08:41:34.640
|bug|status-completed|
<p>I earlier asked <a href="https://electronics.meta.stackexchange.com/q/2675/17592">Dismissing messages</a>, on which Anna Lear <a href="https://electronics.meta.stackexchange.com/a/2685/17592">answered</a> and I commented on that answer. I assume that I don't get a reply because <a href="/questions/tagged/status-completed" class="post-tag moderator-tag" title="show questions tagged &#39;status-completed&#39;" rel="tag">status-completed</a> is included, that's why I ask this new question.</p> <p>Anna Lear:</p> <blockquote> <p>These notification messages should now disappear on their own without any clicking required at all.</p> </blockquote> <p>However, here is a list of messages that still have crosses to close them:</p> <ul> <li>You can accept this answer in x minutes</li> <li>You can't vote for your own post</li> <li>Thanks, we'll take a look at it (after flagging)</li> </ul> <p>Can this be fixed? As Olin Lathrop <a href="https://electronics.meta.stackexchange.com/questions/2675/dismissing-messages#comment4442_2675">points out</a> this is very annoying behaviour.</p>
Messages still have crosses
<p>To clarify, that isn't the <strong>Expanded Hover Card</strong> that <em>Established Users</em> get, it's much more in line with the basic card that everyone has. The avatar list for each review queue is to show who is actively reviewing, so <em>only</em> the avatar is shown by default. </p> <p>The <strong>Expanded Hover Card</strong> is only available for question and answer posts. You'll notice that you still can't see an <strong>Expanded Hover Card</strong> on <em>Non-Established Users</em> question and answer posts. </p>
2698
2013-02-21T15:10:17.167
|support|
<p>PeterJ isn't an <a href="https://electronics.stackexchange.com/privileges/established-user">established</a> user:</p> <p><img src="https://i.stack.imgur.com/QMdYn.png" alt="enter image description here"></p> <p>But on the review page he suddenly is:</p> <p><img src="https://i.stack.imgur.com/KqPw1.png" alt="enter image description here"></p> <p>Why is that?</p> <p>Also: on the review page we see images of the latest reviewers. Established users don't have shadow right there:</p> <p><img src="https://i.stack.imgur.com/q0AhZ.png" alt="enter image description here"></p>
User shows on Review up as established user
<p>When you place a cap, right click, "edit parameters" and choose polarized. It's there already.</p>
2710
2013-02-27T18:06:04.480
|support|circuitlab|schematic|
<p>At first: many thanks for <a href="https://electronics.meta.stackexchange.com/a/2708/17592">embedding CircuitLab</a>! It's cool, easy to use, et cetera, et cetera. Major improvement of the site.</p> <p>Here's a little feature request: <em>can we have an electrolytic capacitor as well?</em> Most of the time, I distinguish electrolytic and ceramic capacitors in my circuits - this increases readability. However, the CircuitLab tool doesn't have an electrolytic capacitor. </p> <p>I hope this is on topic here, but perhaps I should ask CircuitLab directly? Their main editor doesn't have an electrolytic capacitor either.</p>
Can we have an electrolytic capacitor in the schematic editor?
<p>There is a nice little feature with Imgur that it actually stores 3 image sizes</p> <p>Take the example posted. Sure its big (and this is why I use inkscape) but it does the job. </p> <p><img src="https://i.stack.imgur.com/020Wv.png" alt="schematic"></p> <p>However... Imgur permits resizing To original image is: <a href="https://i.stack.imgur.com/020Wv.png" rel="nofollow noreferrer">https://i.stack.imgur.com/020Wv.png</a> inserting a sizing letter (<a href="https://i.stack.imgur.com/020Wvm.png" rel="nofollow noreferrer">https://i.stack.imgur.com/020Wvm.png</a>) accesses different sizes. The downside is you would need to remove the letter to edit the cct.</p> <p><a href="https://api.imgur.com/models/image" rel="nofollow noreferrer">https://api.imgur.com/models/image</a></p> <p><img src="https://i.stack.imgur.com/020Wvm.png" alt="schematic"></p>
2716
2013-02-28T13:01:40.913
|feature-request|circuitlab|
<p>If it's up to me, the CircuitLab schematics can be somewhat smaller. See this example: <a href="https://electronics.stackexchange.com/q/59387/17592">What is the purpose of R2 in this discrete voltage regulator circuit?</a>. </p> <p>I tried earlier to make a small circuit on <a href="https://electronics.stackexchange.com/q/59325/17592">this question</a> but didn't, because it would be way too large.</p> <p>Can the schematics be scaled to a smaller size when it isn't a large circuit? The main site has this feature <a href="https://www.circuitlab.com/forums/feature-requests/topic/c8qncqx8/smaller-exported-images/#comment_4328" rel="nofollow noreferrer">as explained on the CircuitLab forum</a>.</p>
Smaller CircuitLab schematics
<p>As mentioned in a comment, we had a similar <a href="https://ux.meta.stackexchange.com/questions/870/allow-new-users-to-post-mockups">question</a> on ux.se about the Balsamiq editor, and there were three reasons to decline it:</p> <ol> <li>Allowing everybody to use it would mean making ux.se &quot;that site on the internet where you can use Balsamiq for free, with no nag screens.&quot;</li> <li>It would also mean that everybody can post <em>any</em> image, because a) you can include any image in a mockup, and b) even if you couldn't, the image of the mockup is rendered client-side, so we cannot actually trust that an image that <em>says</em> it's a mockup really <em>is</em> a mockup.</li> <li>We would like new users to not rely on this tool too much, because good questions and answers should consist of more than just an image – it should contain explanatory text, with the image as an illustration. Thus it helps to have a tiny bit of familiarity with the Stack Exchange system before starting with mockups.</li> </ol> <p>Now, on electronics.se two of those three are not an issue. As for 1., I got this email from the great people at CircuitLab:</p> <blockquote> <p>I also saw this discussion on Electronics Meta: <a href="https://electronics.meta.stackexchange.com/questions/2717/why-is-the-circuitlab-tool-privileged">Why is the CircuitLab tool privileged?</a></p> <p>We're fine with relaxing the reputation limit for posting a schematic if that's what you and your colleagues at Stack Exchange and the community at Electronics SE ultimately decide. Compared to Balsamiq, CircuitLab is already a free tool, so we have no problem with it on that front.</p> </blockquote> <p>As for 2., due to implementation differences between the Balsamiq integration and the CircuitLab integration, we can actually be sure whether an image really is an image of a schematic.</p> <p>This leaves only 3., and we have decided to give it a try regardless.</p> <p>The schematics editor is now open for everyone, and CircuitLab schematics created through the integrated editor (and only those) are exempt from the &quot;new users cannot post images&quot; restriction. We'll be keeping an eye on it, and if it starts to cause issues, we may undo this change. But we're optimistic that it won't.</p>
2717
2013-02-28T13:19:25.187
|feature-request|status-completed|schematic|circuitlab|privileges|
<p>Recently, the <a href="https://electronics.meta.stackexchange.com/a/2708/17592">CircuitLab tool was added</a>. This tool is privileged so that only people with 11 reputation can post images.</p> <p>I understand <a href="https://electronics.stackexchange.com/privileges/new-user">the idea of privileges</a>:</p> <blockquote> <p>Because we allow participation from anonymous internet users, we must take some precautions to ensure that the rare malicious or spammy anonymous user doesn't ruin the experience for everyone else.</p> </blockquote> <p>But I don't see why the CircuitLab tool should be privileged. I'd say it's very hard to create spam with it. Can anyone explain this to me?</p> <hr> <p>(And no, I do not think it's hard for a new user to get 11 reputation, but I don't see why there has to be a limit.)</p>
Why is the CircuitLab tool privileged?
<p>This isn't you or your mac; sorry, we broke something. Fixed already and deploying ASAP (as in: the next few minutes).</p> <p>See also: <a href="https://meta.stackexchange.com/questions/170029/adding-a-picture-from-the-web-doesnt-work-on-google-chrome">Adding a picture from the web doesn&#39;t work on Google Chrome</a></p>
2725
2013-03-04T12:42:27.487
|bug|status-completed|
<p>I notice that the add-image component dialogue has changed since I last asked a question. I now cannot upload an image using my Mac in either Chrome or Safari, using either an image from my computer or from the web. Is anyone else having this problem?</p> <p>Here is the post in question . . . </p> <p><a href="https://electronics.stackexchange.com/questions/59730/why-are-the-pads-different-for-tht-smt-pads-in-eagle">Why are the pads different for THT / SMT components in Eagle?</a></p>
I'm having trouble adding images to a question using a Mac
<p>A temporary solution is to add a UserScript to your browser, like this:</p> <pre><code>// ==UserScript== // @name Remove dotted border on EE.SE // @namespace http://*electronics.stackexchange.com/* // @version 1.0 // @include http://*electronics.stackexchange.com/* // ==/UserScript== // .gravatar-wrapper-32 is for small pictures like under a question / answer // .gravatar-wrapper-48 is for somewhat bigger pictures like on the users list // .gravatar-wrapper-128 is for the profile picture $('.gravatar-wrapper-32, .gravatar-wrapper-48, .gravatar-wrapper-128').css({ border: '0' }); </code></pre>
2731
2013-03-09T11:55:03.250
|bug|design|
<p>I noticed this:</p> <p><img src="https://i.stack.imgur.com/yCMg6.png" alt="enter image description here"> </p> <p>As you can see, some users have a dotted border around their profile picture. I recently changed my profile picture, but Renan didn't, as far as I know. This really isn't something on the picture, I checked, it's in the stylesheet of the website (see below).</p> <p>I can't find a logic in who has the border and who hasn't. It's the same all the time though. </p> <hr> <p>Users with a dotted border have a somewhat malformed profile page: the user's info starts below the image. Have a look at <a href="https://electronics.stackexchange.com/users/17592/camil-staps">my profile</a> for example.</p> <p><img src="https://i.stack.imgur.com/pFuqX.png" alt="enter image description here"></p> <p><strong>This has been fixed. The dotted borders still appear though.</strong></p> <hr> <p>More users: <a href="https://electronics.stackexchange.com/users/19926/i-rakshith">I rakshith</a>, <a href="https://electronics.stackexchange.com/users/10456/nick-rosencrantz">Nick Rosencrantz</a>, <a href="https://electronics.meta.stackexchange.com/users/19318/jeremy-tunnell">Jeremy Tunnell</a>, <a href="https://electronics.stackexchange.com/users/19941/angelatlarge">angelatlarge</a>, <a href="https://electronics.stackexchange.com/users/11970/tom-l">Tom L.</a>, <a href="https://electronics.stackexchange.com/users/16324/brian-drummond">Brian Drummond</a> (after he changed his avatar, I believe)</p> <hr> <p>This is EE-specific. I can't find any picture on WordPress Answers or TeX with the dotted border.</p> <hr> <p>I took a look at the HTML, and found that the users with dotted border are in a <code>&lt;div&gt;</code> with class <code>gravatar-wrapper-32</code>. The other users are in a <code>&lt;div&gt;</code> with no class at all. The CSS (<code>border: 1px dotted rgb(195, 181, 156);</code>) is linked to the class <code>gravatar-wrapper-32</code>.</p>
Why do some users have a dotted border?
<h2> Mods can not merge accounts anymore </h2> <p>We lost this right because if we merge accounts, incorrect, we can give someone private information about the other. This is a privacy violation, so merge now must go through a more standard channel. You must use the <a href="https://electronics.stackexchange.com/help/user-merge">Merge Function under Help</a>. Take a read through the procedure, it is not very quick, especially for those using browser cookies.</p> <p>We can still do other things, in relation to the link where I talk about dealing with Tony we just destroy all the accounts, instead of merging it back to his account, so you wont be able to tell who was a random user and who was Tony in 6 months, but his content is still deleted.</p> <h2> Want multiple accounts? </h2> <p>You can have as many accounts as you want, I have had an SE employee suggest I get a couple others and use the site from them occasionally so that I know how it looks and feels to others. For a long time I thought that everyone could go back and edit comments indefinitely... Many issues like that arise.</p> <p>A user can have multiple accounts, what you were seeing was probably symptomatic of someone having an account based on a browser cookie and loosing the ability to access that account and having to make a new one.</p> <p>We have an issue if your sock puppets are voting for you and you are voting for them, then it is vote fraud. We also have an issue if you are creating multiple accounts to get around an existing question ban or suspension. </p> <h2>Do not reject edits...</h2> <p>if it is from the same user trying to update their question (I hit the character limit with 0 characters left at the end of that sentence). Of anyone, that use has the most rights to edit their question. For all you know they have contacted SE via email for a merge and are waiting for it, but they don't want to leave a crap question on the site and are trying to improve it. </p> <p>You can reject edits for the normal reasons, but seeing that a user has more then one account is not one of them, previously when I would see this appear I would just fix it, but that is no longer an option, so cut the user some slack.</p> <p>There is no damage done by this, it might look a bit odd, but who cares, most people pay no attention to edits. We just want the best quality answer we can have.</p>
2736
2013-03-10T13:33:50.090
|support|
<p>Yesterday I saw a proposed edit by "gutto" to a <a href="https://electronics.stackexchange.com/q/60334/4512">question</a> asked by "guto". I rejected the edit with a note saying he should use the same account to edit as he did to write the question. He wouldn't need anyone's approval to edit his own question anyway. Also, having the same user speak from different accounts causes confusion.</p> <p>I thought multiple accounts where against the rules, or at least discouraged, and I know mods can merge accounts (I learned this when a particular user that is now in the penalty box for a long time kept creating "sock puppett" accounts, and I heard mods talk about "merging" them to his real account. See <a href="https://electronics.meta.stackexchange.com/questions/1408/suggestion-spend-some-reputation-points-up-front-every-time-you-submit-an-answe#comment3727_1408">fourth comment down</a> by Kortuk.). This guy managed to get his edit thru with the second account anyway, so I flagged it for moderator attention figuring a merge was in order and perhaps a warning to the guy to not create multiple accounts in the future. However, my flag was declined with the cryptic note <i>"Ask in meta about it for more feedback, but it can actually be very hard to recover an account if they loose it. The process is not obvious and I can no longer just merge for them. Do not reject edits"</i>.</p> <p>Now I'm really confused. That note raises more questions than it answers. Something changed recently so that mods can't merge accounts as before? What can they do? When is it appropriate? What really is the policy on multiple accounts? Isn't having the same user, particularly in the same question, using multiple accounts a bad thing? What is meant by recovering a account if someone looses it? What do you mean "loose" it anyway? Note that the edit with the second account was only about 9 hours after the question, so the original couldn't have been considered inactive or something. What is meant by flat out "Do not reject edits"? Isn't that what the edit queue is for? If we're never supposed to reject one, what's the point of having us review edits?</p>
When can mods merge accounts?
<p>Electrical engineering covers much more than electronics. Is the intention of this site to deal with electronics only, or all electrical engineering fields, including such things as:</p> <ul> <li>power generation</li> <li>transimssion</li> <li>telecommunications</li> <li>signal processing</li> <li>industrial automation and controls (SCADA, DCS etc.)</li> </ul> <p>There doesn't seem to be a lot of discussion on non-electronics electrical engineering topics.</p>
2745
2013-03-12T16:38:21.070
|discussion|electronics|
<p><a href="https://meta.stackexchange.com/questions/12952/electrical-engineering-version-of-stack-overflow">This</a> post from 3 years ago seems to show that <a href="https://electronics.stackexchange.com/">https://electronics.stackexchange.com/</a> was originally named Electronics Exchange. Now it is Electrical Engineering. I was wondering why this change was made. I know I didn't give it a proper look when I was looking for a decent electronics site (I assumed it was pure electrical engineering, not electrical <em>and</em> electronic). </p> <p>Was there a reason why it was renamed? I feel like the current name isn't necessarily all-inclusive of the content (nor would Electronics Exchange). Has there been a discussion somewhere (I have searched and can't find it) which led to the renaming or suggested other names?</p> <p><strong>EDIT</strong>: I just saw <a href="https://electronics.meta.stackexchange.com/questions/2733/do-pic24-pic32-questions-belong-here?cb=1">this</a> on the community bulletin and thought it was a perfect case of what I am describing.</p>
Renaming Electrical Engineering
<h2>There is no problem.</h2> <p>I'm from outside the States and had no idea the two weren't the same thing. And even when I would have known <a href="http://in.answers.yahoo.com/question/index?qid=20110323065835AAMjxmy" rel="nofollow noreferrer">the difference</a>, I would've come to this SE site to ask questions about electronics. </p> <h2>SEO</h2> <p>From an SEO (Search Engine Optimalisation) point of view, I'd say this name is really good. We have "Electrical Engineering" as a name, but "electronics" in the URL. That makes this site score high in search engines, both on queries with "electrical" and on queries with "electronics". </p> <p><strong>Name changes aren't nice</strong></p> <p>Not only because 'we're used to electrical engineering', but also because of the search engines. They'll have to get used to the new name, crawl all pages again, et cetera, et cetera. You really don't want to have two different names wandering around the net. In the end, the search engines will get used to the new name though, so this is merely a temporary problem.</p> <h2>The description says it all</h2> <p>We should ward for descriptions in titles. For example, "This is my homepage about blah", isn't a good title, it's a description. The title should've been either "My homepage" (though it doesn't say a lot) or "Blah". </p> <p>EE's description on our <a href="https://electronics.stackexchange.com/faq">FAQ</a> (and <a href="http://stackexchange.com/sites">stackexchange.com/sites</a> as well) says:</p> <blockquote> <p>Q&amp;A for electronics and electrical engineering professionals, students, and enthusiast (...) which include electronics, physical computing, and those working with microcontrollers, Arduinos and embedded systems.</p> </blockquote> <p>That's a very good description and I think anyone can agree with that. We should ward to put the description of the site in the title. We do not want a title like "Electrical <em>and</em> electronics engineering, <em>and</em> a bit about electromagnetic forces <em>and</em> communication schemes as well". </p> <h2>We have a logo</h2> <p>And apart from the title and the description, the logo (and the overall site design) is a good way to profile your website. Our logo is more electronics than electrical engineering:</p> <p><img src="https://i.stack.imgur.com/j8jEF.png" alt="enter image description here"></p> <p>... and thereby shows everyone that electronics is included in the scope of this site. Also the overall site design, with chips, LEDs and a breadboard, is more of electronics than of electrical engineering.</p> <h2>"Electronics" is confusing</h2> <p>As stated in <a href="https://electronics.meta.stackexchange.com/q/149/17592">Possible Name Change?</a>, the word "Electronics" is associated with consumer electronics. This question shows there was a problem with the word "Electronics", and even now, I think we have too much questions about consumer electronics. We really don't want more, do we?</p> <h2>So no,</h2> <p>I don't think we should have a name change.</p> <h2>But yes,</h2> <p>I will be reading this question and its answers with great interest to see with what you come up.</p>
2752
2013-03-12T23:29:06.397
|discussion|
<p><a href="https://electronics.meta.stackexchange.com/questions/2745/renaming-electrical-engineering">Here</a> is another question I just posed, but really what I wanted to drive at was possible suggested name changes to the site. I know a lot of people love the site the way it is so this is just a discussion so I'd like to ask everyone to be constructive here. There's no bad ideas in a brainstorm :)</p> <p>Note that <a href="https://electronics.meta.stackexchange.com/questions/149/possible-name-change">this</a> thread is not a duplicate as it was formed when the site was named something else (however you may want to quickly go through it because there is still relevance there).</p> <p>My primary motivation here is that while 'Electrical Engineering' is all-encompassing in the US, it doesn't mean the same thing in the rest of the world (there is a distinct difference between electronic and electrical outside the States). What I would like to hear suggestions for is a site name that is specific enough to represent the content well but broad enough to capture <em>everything</em> about the site.</p> <p>So does anyone have any ideas?</p>
Should there be a name change to the site?
<p>They are a boundary that is on topic on both sites, as such I feel we should allow users to choose where they ask and keep the questions there.</p>
2754
2013-03-13T08:28:45.210
|discussion|
<p>Questions asking about how HDL constructs relate/imply hardware or vice versa, I think belong on Electronics.</p> <p>Question about testbenches, compiler errors feel like programming questions to me, and StackOverflow already does a good job of answering them, <a href="https://stackoverflow.com/search?tab=newest&amp;q=verilog%20is%3aquestion">Verilog Qs on StackOverflow</a>.</p> <p><strong>Example questions I think could be migrated</strong> :</p> <ul> <li><a href="https://electronics.stackexchange.com/q/60724/13513">Testbench connectivity issues</a></li> <li><a href="https://electronics.stackexchange.com/q/60765/13513">Incorrect ideas on how the simulator works</a></li> <li><a href="https://electronics.stackexchange.com/q/60712/13513">Closed Testbench issue</a></li> </ul> <p>I would like some guidelines on this topic as I see a number of posts which I think would get better answers on StackOverflow, but do not want to unnecessarily/incorrectly flag them.</p> <p><strong>Update</strong> Based on the The Photons feedback I understand that downvotes on meta are against the idea, not the question. From the down votes against migrating questions implies the community welcomes HDL based questions.</p>
Where do HDL (Verilog) questions belong?
<p>This is a meta.SE policy. Links break. We have been around 4 years. I get flags for link only broken answers all the time.</p> <p>Your answer should answer the question and only use links for further reading/reference for material.</p>
2757
2013-03-14T16:17:52.017
|support|hyperlinks|answers|
<p><a href="https://wordpress.meta.stackexchange.com/q/2409/26607">What about answers consisting (almost) only of a link to an external website? - or - 'How do I answer' section for the faq</a></p> <p>Generally, this link from the WordPress SE says that answers just linking to a solution are <em>wrong</em> per the <a href="https://electronics.stackexchange.com/faq#deletion">faq#deletion</a>:</p> <blockquote> <p>Answers that do not fundamentally answer the question may be removed. This includes answers that are … barely more than a link to an external site</p> </blockquote> <p>Those answers should be made a comment if useful or removed if not, so <a href="https://wordpress.meta.stackexchange.com/a/2410/26607">says</a> Rarst ♦. </p> <p>I agree with him for there, but am not quite sure if this is / should be the same here. What about links to Wikipedia, App Notes, datasheets, and other texts or documents that are <em>widely used, maintained by big organizations and generally considered to be accessible in the future as well</em>?</p> <p>For example, look at <a href="https://electronics.stackexchange.com/a/60938/17592">pjc50's answer on Save energy with PIC project</a>. I consider this answer to be useful, though it's nothing more than a "yes" and a link to where the "yes" is explained. Also see <a href="https://electronics.stackexchange.com/q/55545/17592">this</a> question which is basically screaming for app notes, which are given in <a href="https://electronics.stackexchange.com/a/55554/17592">this</a> answer.</p> <p>There's a vague line between "barely more than a link to an external site" and "very useful, official documentation".</p> <p>This meta-question is intended to give support on how to treat answers like the two linked above.</p> <hr> <p>Related: <a href="https://electronics.meta.stackexchange.com/q/44/17592">the problem of answers with links</a></p>
Policy on answers only containing a link
<p>It was migrated, the post there will no longer update for changes here. You will not earn rep for it here and at some point they will clean how and remove it from Meta.SO and you will lose it showing on your account.</p>
2763
2013-03-15T16:30:00.340
|bug|status-bydesign|
<p>I don't use this site all that much, which is evidenced by the fact I asked <a href="https://electronics.meta.stackexchange.com/questions/2745/renaming-electrical-engineering">a question</a> which should have been posted to Electrical Engineering Meta on Meta Stackoverflow.</p> <p>The question now has 7 votes and I have accepted a well-thought out answer. In my <a href="https://meta.stackoverflow.com/users/213966/tmwoods">Meta Stackoverflow</a> user profile summary it shows the question has three upvotes 3 (and my account shows +15 in rep for these 3 votes) and in my <a href="https://electronics.meta.stackexchange.com/users/19530/tmwoods">Electrical Engineering Meta user profile</a> it shows the correct score of 7 on the question (and obviously there is no affect on my reputation). You can probably guess that the question was moved when it had 3 upvotes.</p> <p>The issue is that the question does not show as being marked as correct in either account. And since I, like probably quite a few people on these sites, have a wee bit of OCD in me this drives me nuts.</p> <p>I hope the description makes the issue clear.</p> <p><strong>EDIT</strong>: This is a screen shot from my Meta Stackoverflow account:</p> <p><img src="https://i.stack.imgur.com/G9wjv.png" alt="SO Meta" /></p> <p>You can see the question shows neither the correct score nor as being marked as correct. I'd like it if it either didn't show up in my questions list there, or else reflected the correct state of the question. Either way works.</p> <p>I also just noticed that the question now shows up correctly in my EE Meta.</p>
Migrated questions don't show as being marked correct
<p>You have to be careful what advice or procedures you pull from Stack Overflow. Some of their stuff is legacy and some of it is purely geared towards (relatively) large communities. The meta organization for meta.SO is the way it is (both site <em>and</em> system meta) because it will be painful to change it.</p> <p>Meta is for site business. It's for people to speak their mind, identify issues with the community, and to gauge public opinion. People do this because they have an issue, or are motivated to improve the site.If there is a reward for stating "popular" opinion, and a penalty for saying something unpopular, then meta devolves into a popularity contest. </p> <p>You get points for being an expert on the site, not for being an expert about the site.</p>
2764
2013-03-15T16:43:41.260
|discussion|
<p>I am not disagreeing with the current set up; I am just curious what the motivation was to design the site this way.</p> <p>StackOverflow and Meta StackOverflow have separate reputation scores (you earn different scores on the two different sites) but here at Electrical Engineering Meta your reputation is linked to your account on Electrical Engineering. </p> <p>Why the difference? Was there a particular reason for one going one way and the other one a different way?</p>
Why is there no reputation earned on this site?
<p>Yes and no. </p> <p>You can use <a href="http://data.stackexchange.com/" rel="nofollow noreferrer">Stack Exchange Data Explorer</a> to run arbitrary SQL queries (read-only, of course) against any site in the network. </p> <p>However, Stack Exchange uses Microsoft SQL Server which doesn't support regular expressions. The closest you'll get is some very <a href="http://msdn.microsoft.com/en-us/library/ms187489.aspx" rel="nofollow noreferrer">limited pattern matching</a> (<code>LIKE</code> syntax). Even the simple regular expression you suggest can't be expressed with the limited syntax, though.</p> <p>Stack Exchange does offer a few APIs that you might be able to use to do this: <a href="http://api.stackexchange.com/">http://api.stackexchange.com/</a>. I haven't looked at the API at all, so I'm not sure if it has the same limitation as the Data Explorer (or what is possible with it).</p> <p>You could always query a larger, unfiltered data set from the API(s) and filter it with regular expressions on the client side (i.e. your own application). </p> <p>I agree with Keelan that you should ask this at SO Meta. Data Explorer is cared for by SQL ninjas, so I'm sure they can provide a definite answer/maybe even offer to implement the feature :)</p>
2768
2013-03-17T03:17:30.927
|feature-request|search|
<p>I could have avoided asking a duplicated question had I been able to use <em>extended</em> regular expressions.</p> <p>Duplicated question is <a href="https://electronics.stackexchange.com/questions/61141/what-is-the-origin-of-the-r-in-resistance-measurements?noredirect=1#comment117360_61141">What is the origin of the &quot;r&quot; in resistance measurements?</a> </p> <p>Had I been able to search for answers using an expression akin to:</p> <pre><code>\b[0-9]+R[0-9]+\b </code></pre> <p>then I might have seen the other answer and not added the duplicate question.</p> <p>Is there a way to do that within the Stack Exchange?</p>
How do I search any of the Stack Exchange sites using regular expressions?
<p>Depending on the tags you use, different code syntax highlighting will be used. You selected the <a href="https://electronics.stackexchange.com/questions/tagged/assembly" class="post-tag" title="show questions tagged &#39;assembly&#39;" rel="tag">assembly</a> tag, which has a code syntax style default. It's not exactly assembly because there isn't a code format for that available. </p> <p>I wouldn't expect a "proper" syntax highlighting for your code as that is more of a file format (Intel Hex) rather than a code language (C, ASM, Python, etc...). The syntax highlighting is done using Google Prettify (see <a href="https://electronics.stackexchange.com/editing-help#syntax-highlighting">editing help</a> for more info). </p>
2788
2013-03-24T15:07:22.773
|bug|status-bydesign|
<p>The syntax highlighting of hex files is bad, as shown in <a href="https://electronics.stackexchange.com/q/62112/17592">What PIC processor was this HEX-file meant for?</a>: </p> <p><img src="https://i.stack.imgur.com/4ggl2.png" alt="enter image description here"></p> <p>Can something be done about that?</p>
Hex file formatting
<p>I don't know if it's changed or I just have some issue, but I NEVER see anything but the meta group for a migration option, and the meta group is usually not a good option when I think of migrating.</p> <p>I would suggest a migrate options list that is <strong>sensitive to the reputation</strong> of the person suggesting the migration ON THE SITE THEY SUGGEST MIGRATING TO. I have certainly seen high-ranked moderators on site A who know nothing of the culture on site B send lousy migrations. Someone, even someone with a rep of 101 on site A, who is a 3K, 5K or 10K rep on site B is going to know that "this question belongs on site B", and should be able to suggest that, easily and sensibly. That should cut WAY down on rejections, and ALSO cut way down on questions that are closed because there is no sensible way to suggest migrating them where they need to go.</p> <p>"You can submit a custom flag" Well, no, you can't, FROM THE MIGRATION page. You SHOULD be able to make a suggestion on the migration page, but you can't, actually. You have to take 3 steps back and wonder how much you're wasting your time by making a generic "other" flag to suggest a specific thing (migration) for which there is already a (mostly broken, IMHO) work path.</p> <p>The current number of migrations completely fails to capture the questions that are simply closed without even a migration attempt, when there might have been a good place for them to be answered.</p>
2800
2013-03-26T16:35:01.273
|discussion|feature-request|status-completed|migration|flagging|
<p>On SO, when you flag a question as off topic, you can choose several options to migrate the question to:</p> <p><img src="https://i.stack.imgur.com/fs8KP.png" alt="enter image description here"></p> <p>We don't have that, except for meta:</p> <p><img src="https://i.stack.imgur.com/i6qZl.png" alt="enter image description here"></p> <p>I see some questions that I'd like to see migrated instead of just closed. For example, this one: <a href="https://electronics.stackexchange.com/q/61792/17592">What can I do as a student to be a better candidate for a hardware design job?</a></p> <p>So, can we have more migration options? What should these options include? For now, I'm thinking of SuperUser, StackOverflow and DIY. What are your thoughts?</p>
More migration options
<p>I declined the flag, for the reason stated in the feedback itself.</p> <p>Although it was little more than a link, that was indeed an answer, it was coherent with the topic of the question and - through the link - it was providing with good content.</p> <p>If you don't like it, you have other ways to act before thinking to flag it:</p> <ol> <li><p>Edit it, adding the content that you think might complete it (in this case, a summary of what the video explains). This is a very valuable contribution that you give to the answerer and the site in general, and it also gains you some rep.</p></li> <li><p>Comment to the answer, saying that it should be something more than just the link, and asking to add the content, as descripted in 1.</p></li> <li><p>Downvote: if the answerer doesn't act over your feedback, or if you think that it may misguide the readers, or if you think it's low quality in general - that's your way of informing the community and the user itself of your opinion.</p></li> </ol> <p>NB: I didn't mean to reject your contribute, which is always welcom and often useful, just pointing out that flags should be used when you don't have other ways to fix.</p>
2805
2013-03-27T14:02:37.433
|support|specific-question|flagging|
<p>I flagged <a href="https://electronics.stackexchange.com/a/62560/17592">this answer</a> as "Not an answer" because it's a link-only answer, and <a href="https://electronics.meta.stackexchange.com/a/2758/17592">this</a> is our policy on that and the FAQ states:</p> <blockquote> <p>Answers that do not fundamentally answer the question may be removed. This includes answers that are … barely more than a link to an external site</p> </blockquote> <p>My flag was</p> <blockquote> <p>declined - flags should not be used to indicate technical inaccuracies, or an altogether wrong answer</p> </blockquote> <p>Why? I didn't mean to indicate a technical inaccuracy.</p>
Why was my flag declined?
<p>Diamond moderators decline flags. </p> <p>Any user using the review queue can dispute your flag.</p>
2809
2013-03-27T16:29:31.923
|discussion|
<p>What's the difference between <strong>declined</strong> and <strong>disputed</strong> flags?</p> <p><img src="https://i.stack.imgur.com/GVaKr.png" alt="enter image description here"></p>
What's the difference between declined and disputed flags?
<p>This is the result of a <a href="https://meta.stackexchange.com/a/173895/151157">recent bug fix</a>. Apparently there was a bug that prevented some posts from appearing in the queue. Since the bug was fixed, those posts that should have been in the queue now are.</p>
2811
2013-03-27T23:52:32.130
|support|
<p>I just checked in and notice within the last couple of hours or so that 76(!) post have suddenly appeared in the review queue for first posts. WTF? I looked at the dates on just a few, and they weren't all recent. Obviously I'm missing something about how the system works or is supposed to work. If anyone knows what's going on and how old posts can suddenly show up in the first posts queue, please explain.</p>
76 first posts in review queue?
<p>Just as some additional non-moderator insight while also being respectful of his identity I have seen quite a lot of the troll's 'work'. I assume it's because I'm in a UTC+11 time zone I'm often around at 'odd' times compared to many other users. An example earlier today were three answers posted from a new account that to quote you:</p> <blockquote> <p>seemed remotely plausible (i.e. not entirely gibberish)</p> </blockquote> <p>In fact one was potentially pretty useful, even though not grammatically great. When reviewing the first posts while I had 'suspicions' it was the same person as I'd been advised by Kortuk in chat I treated that individual post on it's merit and it was OK. The following two answers he posted while not great didn't warrant any sort of flag.</p> <p>That was followed by a suggested edit on a question from the same account that included something along the lines of "OP is as much as an idiot as (moderator name)". Then there was a similar sequence from another new account shortly after that followed a similar pattern with the final 'answer' including the OP should stop smoking whatever they are on or similar.</p> <p>I haven't been around long enough to know what the original suspension was about, but considering he does seem to have insight into some topics it's a shame he doesn't concentrate his efforts on constructive contributions instead of endless creation of sockpuppet accounts for trolling.</p>
2826
2013-04-02T21:18:21.220
|discussion|moderation|answers|new-users|
<p>I apologize if thise has been discussed before, but I didn't see anything. There has been a rash of new user acounts answering questions, and an equal (rush) to protect them </p> <ul> <li><a href="https://electronics.stackexchange.com/questions/64170/2-1-sound-system-noise">2.1 sound system noise</a> <a href="https://electronics.stackexchange.com/users/22024/user22024">user22024</a></li> <li><a href="https://electronics.stackexchange.com/questions/64066/bus-vs-seperate-return-paths-for-3-phase-bridge">Bus vs. seperate return paths for 3 phase bridge?</a> <a href="https://electronics.stackexchange.com/users/22023/user22023">user22023</a></li> <li><a href="https://electronics.stackexchange.com/questions/64083/high-to-low-impedance-microphone">High to low impedance microphone</a> <a href="https://electronics.stackexchange.com/users/22020/user22020">user22020</a></li> <li><a href="https://electronics.stackexchange.com/questions/64078/mixing-ac-and-dc-power-in-a-circuit">Mixing AC and DC power in a circuit</a> <a href="https://electronics.stackexchange.com/users/22019/user22019">user22019</a></li> </ul> <p>And so on. As of this writing all these users can be seen on the main page next to the questions they posted answers to, but now the questions are protected, and the new accounts are suspended. The answers seemed remotely plausible (i.e. not entirely gibberish) though the questions were out of my area of expertise for sure.</p> <p>On the <a href="https://electronics.stackexchange.com/questions/63967/pull-switches-why-so-expensive">pull switches</a> question the new user account posted</p> <blockquote> <p>Whatever you choose, the contacts must be gold plated to prevent corrosion.</p> </blockquote> <p>On one hand, this answer <em>could</em> make sense: gold protects from corrosion. On the other hand, in this case the OP specifically said that corrosion is not a factor.</p> <p>What is going on?</p>
Many new user accounts+answers+suspensions and protected questions
<p>It's a 5 minute limit. This has been discussed many times on Meta.StackOverflow if you want to know more about it. </p> <p>I believe one of the staff said that if they logged every single edit as a separate revision it would add a huge load on their database, as most people reread their posts and edit typos immediately after posting. </p>
2828
2013-04-03T11:24:40.310
|support|
<p>In reviewing a suggested edit I made some further improvements and then realised moments later I'd made an incorrect assumption that the OP was using a pair of microcontrollers, considering it was a CAN bus it could have involved more than two devices so I removed my reference to a pair. It is edit three on the following revision list:</p> <p><a href="https://electronics.stackexchange.com/posts/64220/revisions">https://electronics.stackexchange.com/posts/64220/revisions</a></p> <p>Apart from my misspelling of 'possiblly' in the edit summary it looks a bit odd now because my first edit has been rolled into the second but using the second edit summary where I'd just removed a "pair of" microcontrollers. No big deal but I wondered for future reference how long you get to re-edit a post without it appearing as a distinct edit?</p>
Time limit before edits commit to revision history
<h1>Moderators</h1> <p>I took the locations of the profile pages<sup>1</sup>:</p> <ul> <li><a href="https://electronics.stackexchange.com/users/7348/clabacchio">clabacchio</a> - Pesaro, Italy - UTC+1</li> <li><a href="https://electronics.stackexchange.com/users/857/kevin-vermeer">Kevin Vermeer</a> - Grand Rapids, MI - UTC-5</li> <li><a href="https://electronics.stackexchange.com/users/230/kortuk">Kortuk</a> - Oklahoma - UTC-6</li> <li><a href="https://electronics.stackexchange.com/users/411/markrages">markrages</a> - US - unknown</li> <li><a href="https://electronics.stackexchange.com/users/638/w5vo">W5VO</a> - Arkansas - UTC-6</li> </ul> <blockquote> <p><sup>1: I hope I don't mess with privacy, that was (and is and will be) never the intention. I thought I could list the information as it was freely available on the <a href="https://electronics.stackexchange.com/users?tab=moderators">moderators tab</a> of the users list. </sup></p> </blockquote> <p>To give some graphical impression, I made this. I estimated one usually is online from ~8:00 am to ~8:00 pm local time. But as that assumption probably isn't right at all, the image is only a "<a href="http://en.wikipedia.org/wiki/Allegory_of_the_Cave" rel="nofollow noreferrer">shadow of reality</a>":</p> <p><img src="https://i.stack.imgur.com/tmrsQ.png" alt="enter image description here"></p> <h1>All users</h1> <p>I made <a href="http://data.stackexchange.com/electronics/query/106848?LastMonths=1#graph" rel="nofollow noreferrer">a query</a> for this. For the last month, the site seems to be the most active from 14:00 to 20:00. <em>I</em> think <em>this is UTC, but I'm not sure.</em></p> <p><img src="https://i.stack.imgur.com/54DPM.png" alt="enter image description here"></p> <p>The last 10 months give a smoother view:</p> <p><img src="https://i.stack.imgur.com/l4hMv.png" alt="enter image description here"></p>
2840
2013-04-06T08:08:56.780
|support|election|moderators|
<p>I encounter a delay in the review of flags when it's morning here in the Netherlands, let's say 6:00 UTC. That made me wonder what the timezones of the moderators are.</p> <p>Other sites use this information when electing new moderators, so I thought I could give it a meta question.</p> <p>Of course, this data is little useful when not compared to the activity of the whole site, so I'd like to see activity stats of the whole site as well.</p>
When are the moderators / is the site active?
<p>That was our fault. When we convert an answer to a comment we get a nice little check box to migrate all comments to that answer also, as sometimes the OP will get in a back and forth discussion there. When the answer was converted to comment your comment was migrated also.</p> <p>All fixed now. Your comment is still on the users answer that was deleted and it will notify them there. </p>
2850
2013-04-10T12:20:50.853
|support|
<p>Earlier today I flagged an answer as "not an answer" and it's since been moved to a comment along with my original comment under the answer. The question I'm discussing is:</p> <p><a href="https://electronics.stackexchange.com/questions/42911/how-do-i-exchange-lcd-display-with-7-segment-leds">How do I exchange LCD display with 7-segment LED's?</a></p> <p>In case anything changes in the comments area the 'answer' was:</p> <blockquote> <p>during 1996 l have found on an electronics magazine(probably'electronics for you' or elektor,i don't rember) on a project of an electronic"lcd clock"based on a wrist watch. where the signals for lcd display were amplified by using some pairs of SL100and SK-100 to drive 3&amp;1/2 LED display connected in matix arrangement. I am still searching for the project.</p> </blockquote> <p>During the first post review process I made the following comment and flagged it:</p> <blockquote> <p>Welcome to the EE.SE! Until you locate the project and can provide further information though you shouldn't really post this as an answer. Once you have enough reputation it would be suitable as a comment but for the moment it would be better to hold-off until you've found the solution that answers the question.</p> </blockquote> <p>Now I'm wondering if I delete my comment if the user will still receive any feedback if they haven't happened to be on the site since? I thought I'd leave the comment for the moment in case it means they won't get feedback but at the same time it may confuse the OP that it's somehow directed at them.</p>
Comments when "not an answer" is processed
<ol> <li>Create Google account.</li> <li>Use Google account to log in to EE with Open ID.</li> <li>See <a href="https://english.stackexchange.com/help/user-merge">this page</a>.</li> </ol>
2861
2013-04-16T14:13:29.747
|bug|website|openid|
<p>I was first going to ask this on Meta, but that was not possible because it required me to do "finish signup" which doesn't work for me. I always have this thing at the top of the screen next to my name that says "finish signup". <img src="https://i.stack.imgur.com/Yvke9.png" alt="Finish Signup"></p> <p>So, I click on it, and see what happens. I get to this screen, and I enter a password.</p> <p><img src="https://i.stack.imgur.com/abYbJ.jpg" alt="Entering Password"> </p> <p>After I do that, it redirects me back to the electrical engineering home and it still says "finish signup". What is wrong with this, and Why can't I sign up, I think I have enough reputation. I wish to be able to vote, so this is why I want to finish signup.</p>
Why doesn't finish signup work?
<p>Stack Exchange works primarily from VC funding. It raised 12 million USD in March' 11. This is what keeps the sites up. </p> <p>The earnings from sources mentioned in <a href="https://meta.stackexchange.com/a/79448/200868">Jeff's answer</a> make up a portion of the earnings but alone, are not enough to keep the servers running and employees paid*. </p> <p>Also, note that the company has expanded significantly since 2011 whereas ads have not increased in number. Stack Exchange is focusing mainly on expanding Careers and increasing earnings from it.</p> <p><sub>*I will add a source for this when I find it. I think it was mentioned in one of the podcasts. </sub> </p>
2878
2013-04-20T12:58:59.037
|discussion|
<p>I am curious as to who pays the bills to keep the lights on around here? Is the site supported purely by advertisements? Does stackexchange itself own the site?</p>
Who pays the bills?
<p>There are already methods in place for Qbans/Abans. </p> <p>They dont release the methods, as it would allow gaming, but a user who receives downvote/deleted questions/spam flags will be blocked from posting further until they modify their posts to have some of the previous actions reversed.</p> <p>Having a pre-review step does not make sense to me, if they are spamming, lets just block them all together from posting. </p>
2884
2013-04-22T18:59:32.503
|feature-request|new-users|spamming|
<p>Last night, while the mods were getting some much-needed rest, a new user started posting advertizements for a blog. The community moderation worked, with many of the posts were marked as SPAM. However, a new user is free (as far as I know) to continue posting spam.</p> <p>I would like to propose that new users who had, let's say three posts marked as SPAM by the community (i.e. 5 SPAM flags) are automatically suspended pending moderator review and/or their posts are not published until they pass review (as <a href="https://electronics.meta.stackexchange.com/users/8627/jippie">jippie</a> <a href="https://electronics.meta.stackexchange.com/questions/2884/proposed-change-automatic-suspension-of-users-who-spam#comment6262_2884">suggests</a>). The moderator review of suspension/required review status might be useful if we are worried about some number of users colluding in chat to gang up on an unsuspecting new user. Note that I am not proposing to ban the source IP: rather just suspend the user account. Yes, the spammer can create a new account and continue defacing the site, however, this is a bit of an extra hoop the spammer has to jump through at no cost to the community, under the present proposal.</p> <p>BTW, a similar question has been discussed in the post <a href="https://electronics.meta.stackexchange.com/questions/645/how-do-we-avoid-spam-recidivism">How do we avoid spam recidivism?</a> here, however, both the problem and the solution proposed are somewhat different.</p>
Proposed change: automatic suspension/review of users who spam
<p>The answer is "Yes". The negative rep from downvotes goes straight into your rep.</p> <p>Nonetheless, given that a downvote is -2 whereas an upvote is +10 (answer) /+5 (question) /+15 (accept) it takes much longer to loose all that rep than to gain it. Almost certainly you'll just be banned before you get down to 1. For an example, see <a href="https://electronics.meta.stackexchange.com/questions/2826/many-new-user-accountsanswerssuspensions-and-protected-questions">this question</a>.</p> <p>One note: on <a href="http://meta.electronics.stackexchange.com">[ee.meta]</a> there is no reputation separate from <a href="http://electronics.stackexchange.com">[ee.se]</a>. The same is not true of <a href="http://meta.stackoverflow.com">[so.meta]</a>, which has its own rep, being kind of the ubermeta for all stackexchange sites.</p>
2887
2013-04-25T02:03:12.837
|discussion|reputation|
<p>Lets say someone works hard and asks on-topic questions to gain a high reputation, then when they achieve that, they troll around the whole site. Can they lose reputation because of this?</p>
Can people lose reputation?
<p><a href="https://electronics.stackexchange.com/faq#reputation">From the FAQ:</a></p> <blockquote> <p>... You can earn a maximum of 200 reputation per day. Please note that votes for posts marked “community wiki” do not generate any reputation, while accepted answers and bounty awards are not subject to the daily reputation limit.</p> </blockquote> <p>This means that (assuming you don't have any questions upvoted) you can receive 20 answer upvotes in a day before the reputation cap kicks in for that day. Other sources of reputation, such as an answer being accepted, or receiving a bounty, do <em>not</em> count towards your 200 reputation limit. </p> <p>Below is a picture of your reputation for April 5th, when you received 230 reputation. You can see each row labeled <strong>upvote</strong> adds up to a total of 200 reputation, which is exactly the reputation cap. You didn't "lose" any reputation due to having more than 20 upvotes. You also had two answers accepted as "correct". These are shown as "+15" and are outlined in green. The "accept" reward does not count towards your 200 reputation limit, nor does reaching the 200 reputation limit prevent you from receiving that 30 reputation.</p> <p><img src="https://i.stack.imgur.com/y3ASU.png" alt="April 5th Reputation"></p> <p>On May 3rd, you received reputation only from upvotes (no "accepts"). You can see that the first 200 reputation was counted, and the other upvotes are acknowledged but not counted. You can still gain reputation, but it can only be from answers accepting, or bounties, not upvotes.</p> <p><img src="https://i.stack.imgur.com/upki0.png" alt="May 3rd Reputation"></p>
2904
2013-05-03T20:28:48.513
|discussion|status-bydesign|
<p>I noticed in the last few hours that my reputation has stopped incrementing when I get new up-votes: -</p> <p><img src="https://i.stack.imgur.com/ALgyq.jpg" alt="enter image description here"></p> <p>Has something gone wrong or has the well dried up?</p>
reputation stopped incrementing