17
\$\begingroup\$

With some recent silliness in meta referring to guidelines for repair questions, I went to our help center to actually look for those guidelines.

Bottom line: I couldn't find any.

If we're going to be sticklers about repair questions, there needs to be some guidance for new users that appears somewhere other than deep in meta -- if for no other reason than to try to minimize the number of meta discussions that a perfectly valid repair question generates!

Should we change our "what not to ask" guidance? Perhaps added a phrase on the consumer devices description that says "or the repair of such devices" would be a good start?

\$\endgroup\$
  • 3
    \$\begingroup\$ I like all the thoughtful responses, but I'm having a hard time turning them into action items. \$\endgroup\$ – Scott Seidman Sep 12 '18 at 14:25
  • 1
    \$\begingroup\$ Just thought I'd mention - some new members manage to post reasonably decent "repair questions" like this one, although the title could use some work. Granted it's far easier to post a decent repair question when you already have the schematic, but I think it only goes to illustrate that at least some attempt at a schematic is a prerequisite for an answerable repair. \$\endgroup\$ – brhans Sep 29 '18 at 14:11
  • 1
    \$\begingroup\$ Then there's the nicely cropped & perfectly focused photos, taking measurements as requested in the comments and posting the results in the question (at least partly). I'd give that repair question an upvote if it just had a decent title and all the measurements were edited in ... \$\endgroup\$ – brhans Sep 29 '18 at 14:12
10
\$\begingroup\$

I frequently leave the following comment (or some variation of it) on repair questions:

Welcome to EE.SE! This appears to be a reverse engineering, modification, or repair question. Please be aware that such questions must involve specific troubleshooting steps and demonstrate a good understanding of the underlying design of the device being discussed, so that you can ask specific, focused questions that can be answered concisely. Otherwise, the question is far too broad. More information can be found here: Is asking how to fix a faulty circuit on topic?.

\$\endgroup\$
  • \$\begingroup\$ Hmm.. It might not hurt to update the post you're linking to, as the "Too Localized" close reason doesn't exist. I'll take a look at it later. \$\endgroup\$ – W5VO Sep 10 '18 at 15:02
  • 2
    \$\begingroup\$ @W5VO: for me the main reason of any kind of debug question being bad for SE is that it involves a forth and back of information because its basically impossible to provide everything up front, because if they could they could usually make the step to solve the problem themselves. \$\endgroup\$ – PlasmaHH Sep 10 '18 at 15:14
  • \$\begingroup\$ So you mean if they have to ask, don't ask... \$\endgroup\$ – Passerby Sep 10 '18 at 17:02
  • \$\begingroup\$ Also, how are you going to dump mod and reverse engineering questions with repair questions, without a community vote on that. That's the first time I've seen those being mixed in like that. \$\endgroup\$ – Passerby Sep 10 '18 at 17:09
  • 4
    \$\begingroup\$ @PlasmaHH -- I disagree that a back and forth nature generally makes a bad question. There's a lot to be learned by observers of that back and forth. \$\endgroup\$ – Scott Seidman Sep 10 '18 at 17:22
  • 4
    \$\begingroup\$ @scott it maybe good for a forum, but not for the single question single answer format of se \$\endgroup\$ – PlasmaHH Sep 10 '18 at 18:17
  • 2
    \$\begingroup\$ @plasma so any question with comments should be banned then. \$\endgroup\$ – Passerby Sep 10 '18 at 18:44
  • 7
    \$\begingroup\$ @Passerby: Don't be ridiculous. Comments serve to help the OP improve the question (or answer). They really aren't meant to be part of the permanent record, or the only source of clarifying information. \$\endgroup\$ – Dave Tweed Sep 10 '18 at 18:47
  • 1
    \$\begingroup\$ @Passerby: I'm not proposing a total ban on such questions, only clarifying the conditions under which they make sense for this site. \$\endgroup\$ – Dave Tweed Sep 10 '18 at 18:50
  • 10
    \$\begingroup\$ @Passerby Exaggerating others' points of view won't help coming to an agreement. I think another way of stating Plasma's position is that while comments are useful to clarify questions when needed, a lot of these will require a back and forth in the comments to extract the needed information. \$\endgroup\$ – W5VO Sep 10 '18 at 18:58
  • 4
    \$\begingroup\$ Might I interject with an experience on another SE community. Long story short, the back and forth required to help me solve my problem occured in the chat and once the back and forth was done, the Original Answer edited his answer to take our back and forth into account without poluting the rest of the original post. Isn't that kind of guideline something to consider? I was extremely pleased with the help I had and I don't think anybody would be angry at the end result. This is the question: stackoverflow.com/questions/50801120/… \$\endgroup\$ – Simon Marcoux Sep 12 '18 at 16:36
  • 1
    \$\begingroup\$ @SimonMarcoux Although I appreciate the existance of the chat facility, for various reasons, I don't use it. Not only on SE, I mean. So I really prefer the back and forth stream of comments that leads to clarifications (that's the main point of comments being allowed!). The big problem is that sometimes the information are not "backported" into the question by the OP and remain in the comments. Thus comments become part of the question and this is bad. There is a solution, though: to step over the OP and editing his question with the relevant information if he is not willing to do so. \$\endgroup\$ – Lorenzo Donati Sep 23 '18 at 10:56
6
\$\begingroup\$

This is the only thing I could find on the subject:

What topics can I ask about here?

This site is for electronics and electrical engineering professionals, students, and enthusiasts. We ask and answer questions about electrical and electronics engineering topics, which include electronics, physical computing, and those working with microcontrollers, Arduinos and embedded systems. We feel the best Electronics Design questions have a schematic, links to pertinent datasheets or some source code in them, but if your question generally covers …

a specific electronics design problem the theory and simulation of electromagnetic forces communication scheme the writing of firmware for bare-metal or RTOS applications

and it is not about …

a shopping or buying recommendation

consumer electronics such as media players, cell phones or smart phones, except when designing these products or modifying their electronics for other uses

Programming software for a PC

Source: https://electronics.stackexchange.com/help/on-topic

enter image description here

The link takes you here where the guidelines on repair questions were decided. Is asking on how to fix a faulty circuit on topic?

Most of the time moderation queue reviewers can't tell the difference between a consumer product and a hobby project or repair of an oscilloscope, maybe that can be clarified in some way.

The part queue reviewers get hung up on is the "good understanding of the underlying design", for a lot of the queue reviewers the logic might go something like this: "This person's question is dumb and is obvious, so they must not have a good understanding, I'm closing this question" at any rate, there might be a way to change the wording in the moderation queue to clarify that. Queue reviewers could definitely be more lenient in repair question cases and other ways.

On the flip side, if people are repairing something, they should document what they have. Too many times an OP comes and it takes a lot of discussion to get their question clarified\documented up to a point where it is answerable, or that process never happens and the question remains on the site in a half baked state.

\$\endgroup\$
  • 2
    \$\begingroup\$ The consumer product part is about asking how to use, or general product questions. Repair even of consumer products are okay, given it is subject to some effort. The part of good understanding being subject to a You must know the answer before you post attitude is spot on. \$\endgroup\$ – Passerby Sep 10 '18 at 18:43
  • 4
    \$\begingroup\$ This is a site fore electronic enthusiasts, students and engineers, who want to communicate. Not random people that walk in off the street and need everyone to hold their hand. If they write a good intelligent question, which is recommended by the guidelines and by the SE.meta, then I support that. But we also need quality here or it will turn into a site like eevblog, thats why stack exchange has a review que and they don't allow anyone to post anything, there are guidelines. We can do better at reviewing, and some opinions need to be changed, we can talk about it if clarity is needed here \$\endgroup\$ – laptop2d Sep 11 '18 at 3:51
  • 1
    \$\begingroup\$ I think you have a point here, namely that the on-topic page could do with a complete overhaul. It needs more details. \$\endgroup\$ – Lundin Sep 20 '18 at 14:02
4
\$\begingroup\$

First you need to start by addressing people who say "repair questions are off topic" ignoring the help center on topic description like a certain mod. And people who think modification questions are off topic.

Second, you need to address people who think the level of information that a repair question needs is "if you don't know how to fix it, then don't ask." That type of circular, catch 22 logic has no place here.

We can all agree that questions that consist of "here's a picture of an enclosure, help fix" are bad. But a repair question where op has taken the part apart and located a possible trouble spot, typically a burn mark or something, and a description of what they did, is good and should not be dismissed off hand simply because it mentions repair.

\$\endgroup\$
  • 2
    \$\begingroup\$ For example here's a question where the OP took the device apart, located the trouble spot, described what they did. Is it a keeper by your standards? \$\endgroup\$ – Dmitry Grigoryev Sep 17 '18 at 13:53
  • \$\begingroup\$ @Dmitry are you saying it (a four year old question with good answers) shouldn't by your standards? Why? \$\endgroup\$ – Passerby Sep 17 '18 at 14:11
  • 3
    \$\begingroup\$ You don't have to put words in my mouth, or answer a question with a question. \$\endgroup\$ – Dmitry Grigoryev Sep 17 '18 at 14:22
  • \$\begingroup\$ If your bringing a four year question up, then it's likely you disagree with it. Why else phrase it like you did? \$\endgroup\$ – Passerby Sep 17 '18 at 14:26
  • \$\begingroup\$ I don't necessarily disagree, trough this question ended up closed by users I respect and a mod. I'm afraid that allowing repair questions without clear rules can result in a rush of questions from folk who don't know how to change a fuse or a battery, and so far your last paragraph suggests that those should be accepted as well. I thought that it would be easier to discuss the rules applied to an example such as the question above. \$\endgroup\$ – Dmitry Grigoryev Sep 21 '18 at 13:29
  • 1
    \$\begingroup\$ We have a question, by Olin, that covers how to chose a power supply. It's both good for newbies and still extensive. Questions that ask such basic things are closed as a dupe of that. While no one is asking how do I change a fuse, if they did, That do can be closed as a dupe of another, a basic how to solder q and a. Or a question should be created on how fuses work and how to replace/chose one. The repair closure is supposed to address questions with ZERO effort or info, like "my tv is broke, fix it", not questions like the one you link which is good, from a hobbyist with info. @dmitry \$\endgroup\$ – Passerby Sep 21 '18 at 15:04
  • 1
    \$\begingroup\$ @Dmitry it is not supposed be used as a "if you don't know, don't ask" catchall closure. And questions do not need to be 1, specifically answerable. Sometimes the root cause isnt known, it's okay to answer with good general but relevant info. That's how electronics work sometimes. Or 2, immediately answerable. A question where someone doesn't know the answer right now does not mean the question should be closed. \$\endgroup\$ – Passerby Sep 21 '18 at 15:13
3
\$\begingroup\$

IMO a reasonable criterion for repair questions would be to require the user to find a specific anomaly they could measure: an unexpected voltage at a test point, a (lack of) continuity which disagrees with the schematic, excessive current, etc. This is what I understand by "involve specific troubleshooting steps and demonstrate a good understanding of the underlying design" which is displayed in the custom close message.

I'm convinced that users who are willing to make a measurement and have a basic understanding of what they are measuring (not just a list of voltages across every accessible component) have a good chance of asking good questions and learn from the answers. Accepting requests to diagnose a broken device by photo can be easily abused and is unlikely to teach the OP electronics.

Yes, it's a dumb heuristic criterion, but I believe that complex elaborated criteria end up being too subjective and thus unenforceable in practice.

And yes, that means that there should be a minimum level of understanding before asking. And yes, it's catch 22 for user who haven't yet made contact with electrical engineering. I think it's justified because frankly I'don't think they will learn anything from answers which tell them which components to replace, and it's not the site's mission to save someone $10 by spending a couple of man-hours figuring out an electrical defect by photo.

\$\endgroup\$
1
\$\begingroup\$

Maybe something like this:

  • The OP has at least some basic knowledge of electronics, and have access to basic tools that are necessary for the repair. Such as for example a multi-meter, oscilloscope, soldering iron.
  • The post shows the research effort so far, with pictures. Schematics or datasheets should be provided if possible.
  • The repair must be about a specific and narrowed-down problem, such as "how do I replace this electrolyte capacitor and what parameters are important to consider" or "what is the best way to de-solder this QFP", rather than "how do I repair my TV".
  • In general, questions about methods of how to best carry out a repair of an electronic device, in terms of best soldering practices, ESD protection, safely disconnecting line voltages etc, are on-topic.
\$\endgroup\$
  • \$\begingroup\$ Regarding the 2nd bullet. We require "demonstrate a good understanding of the underlying design". In practice that requires the O.P. to have the schematic. (We have examples where a repair question remained open, because it had a schematic from a manual.) Old equipment came with schematics in user manuals or repair manuals. Today's equipment doesn't come with schematics, except when it's based on open source hardware. Does that mean that majority of the repair questions don't clear this guideline? Yes. Does that mean that we should relax that guideline? No. \$\endgroup\$ – Nick Alexeev Sep 21 '18 at 21:11
  • 1
    \$\begingroup\$ That's a problematic interpretation of the rule :/ \$\endgroup\$ – Passerby Sep 23 '18 at 4:00
  • \$\begingroup\$ Oscilloscopes are not tools usually available to hobbyists (especially people with low income or in developing countries). If that had been a criterion, I would never have done anything in electronics. I've got my first oscope after about 15 years "of electronics" (I started when I was 12yo) and well after I passed my 2nd or 3rd exam in electronics at the uni. So, sorry, NO! I would strongly vouch against any heavily income-related filter criterion. At least until this site is marketed as "not only for professionals". Then I could agree, but I could also reconsider my future contributions. \$\endgroup\$ – Lorenzo Donati Sep 23 '18 at 11:09
  • 1
    \$\begingroup\$ @LorenzoDonati It's not about income or professionalism, but simply about the ability to actually carry out the repairs. If someone has a question regarding timing, various data buses and so on, they simply can't trouble-shoot it without a scope. So what tools that are required depends entirely on the nature of the question. \$\endgroup\$ – Lundin Sep 24 '18 at 6:58
  • 1
    \$\begingroup\$ @Lundin I totally agree with you about the need of some specific/advanced tools to troubleshoot some issues. So I'm OK with telling the OP that they can't effectively solve the problem without them, and explaining why. This will teach them something. I'm not OK with closing a question just because the OP doesn't know upfront he needs more advanced tools (and which ones). If they come here for help and show effort I think we owe them at least an explanation on what they need to try to solve the problem. ... \$\endgroup\$ – Lorenzo Donati Sep 24 '18 at 18:48
  • \$\begingroup\$ @Lundin ... Moreover, sometimes some problems could be solved by a more heuristic approach, without more advanced tools, perhaps by wasting more time in the process and with some trial and error. These "McGyver-style" techniques are sometimes even useful on the field and are often fun (which is the ultimate goal for the hobbyist, anyway). \$\endgroup\$ – Lorenzo Donati Sep 24 '18 at 18:48
  • \$\begingroup\$ An interesting corollary is maybe we need to split up actual repair from identifying the problem \$\endgroup\$ – Scott Seidman Sep 24 '18 at 21:54
  • 1
    \$\begingroup\$ @LorenzoDonati The point here is that if someone is surprised that they need a volt meter to measure voltage, or that they need a soldering iron to repair a PCB, the question cannot possibly be within the topic of electrical engineering. The site changed name from Electronics to Electrical Engineering for a reason, and it was to avoid questions from people with zero domain knowledge. \$\endgroup\$ – Lundin Sep 25 '18 at 6:32
  • 1
    \$\begingroup\$ @Lundin Fair point, and I agree. That's why I objected in the first place: an oscilloscope is a whole different beast than a soldering iron or a multimeter. Someone can have far more knowledge than zero-level and still not having access to an oscilloscope (or other high-end equipment). \$\endgroup\$ – Lorenzo Donati Sep 25 '18 at 8:19
  • \$\begingroup\$ @ScottSeidman That's an interesting view. I can't decide right now for a general policy in that sense, but I admit that splitting detective work from repair procedure may be worthwhile in many cases. In fact I dare say there are some cases in which one of the two is the main problem, whereas the other is trivial. \$\endgroup\$ – Lorenzo Donati Sep 25 '18 at 8:22
-3
\$\begingroup\$

Yes, some good responses and comments.

Here again I get that horrible feeling in my gut thinking what if I was the one asking. I enjoy sharing my knowledge on SE to help people. I hate asking questions because my questions are generally (I feel) harder and those I cannot answer with simple searching and so would like to turn to specialists for help. I am reluctant to ask, not because I think I will not get an answer but because I will be judged on my style, location or format more than the merits of the question.

I have run afoul of mods on some stacks that insist that their way is the only way even though it does not serve the true goals of SE. I believe this goal to be the storing of the collected knowledge of humanity in preparation for seeding an AI on an interstellar colony ship to help with unexpected things that need solving.

Fixing a thing is the epitome of something that needs solving. Suggesting a trip to radio Shack (RIP) or Amazon when on a one way trip is impossible, the same goes for a disadvantaged citizen of earth that has to choose between a working oven and food to put in the oven.

@Scott Seidman you were hoping for concrete suggestions, here are three.

  • SE should eliminate close voting and bad question shaming with immediate effect (they only serve the ego, there are no other limitations that are aproacing from having poor questions).
  • SE should add tags for repair, homework, new member that are easy to set. A further suggestion would be to make marking as duplicate super simple for high rep users, I just had a look and again was unable to find the button for it. Also make the duplicate marking feature search for likely suggestions and list them by various options with those that the person has been involved in personally (Q, tag, A, Comment, vote, reopen etc) as those are the questions they remember best.
  • SE should implement a system that hides poor (repair) questions from critical members so they are not required to see them as soon as their quality becomes apparent.

The rest of the system could work as before with those that want to answer questions doing so (gain points where others would close a question) and those that hate newbies and homework and repair and children and spam never having to see it so enjoying SE more and not having a need to criticize.

I have made these suggestions before and the general response has been they will not be implemented. That they result in poor data. That most people find questions here outside of search. That questions with no or poor answers should not even be here. That mods and long term members don't like to wade through rubbish questions (they would rather take the trouble to close vote them than just click next). That a visitor having found their way to a question without an answer will cause that visitor to suffer PTSD rather than receive some enlightenment in comments, at least a sense of how difficult a question they have posed.

These objections are spurious as my suggestions pretty much eliminate the objections simultaneously with making SE a better place to store information and find this information.

Bandwidth and disk space are no longer even an excuse to use for limiting any question. Have the quality of the question determine the visibility but not the existence.

Remember SE needs to evolve to serve the public or it will eventually stagnate and then serve no purpose even for the owners who are anything but transparent in their rule making. Their stated goals are nice enough for them but there is a lot of manpower that they get for free that they take for granted and do not always treat with respect.

\$\endgroup\$
  • \$\begingroup\$ One of the biggest services SE provides is teaching people to ask good questions. \$\endgroup\$ – Scott Seidman Oct 14 '18 at 19:45
  • 1
    \$\begingroup\$ @ScottSeidman Is this a stated goal or just some lame justification for otherwise bullying visitors? \$\endgroup\$ – KalleMP Oct 14 '18 at 20:22
  • 1
    \$\begingroup\$ Good questions are a stated goal. You can close and delete bad questions, or you can teach those who post bad questions to ask good questions. \$\endgroup\$ – Scott Seidman Oct 15 '18 at 12:58
  • \$\begingroup\$ Scott I cannot get my head around the fear of bad questions, disk space is cheap and they can be personally or automatically ignored. The reason I think the method of training the OPs to ask 'good' questions is wrong and just pointless is because they are usually first time visitors and if they are treated badly they are last time visitors (I hate asking questions as I know what I am letting myself in for in asking in a field that I am not a knowledgeble in, but those are the fields I need help with) and this does not increase the data set which should be the stated goal. \$\endgroup\$ – KalleMP Oct 16 '18 at 19:51
  • \$\begingroup\$ Hmm, already 3 down votes on Meta (as usual) just to remind me that top down control is the order of the day on a volunteer operated site. Almost every time I answer something on Meta I am left feeling violated like I should just shut up and leave SE, it should not work like that, there should be a desire to help people and not to bully them. There should be a genuine attempt to develop software tools to make the frustrations go away instead of codifying social control that does not serve any useful purpose. \$\endgroup\$ – KalleMP Oct 16 '18 at 19:56
  • \$\begingroup\$ Scott I cannot get my head around the fear of bad questions, disk space is cheap and they can be personally or automatically ignored. The reason I think the method of training the OPs to ask 'good' questions is wrong and just pointless is because they are usually first time visitors and if they are treated badly they are last time visitors (I hate asking questions as I know what I am letting myself in for in asking in a stack that I am not knowledgeable in, but those are the fields I need help with) and this does not increase the data set which should be the stated goal. \$\endgroup\$ – KalleMP Oct 16 '18 at 19:57
  • \$\begingroup\$ I belong to a conservation organization that does things a certain way. I griped about not liking some aspects of how they do things, but liking other aspects. Somebody told me "That other conservation organization does things the way you like, and I donate money to it". I got the point. I still participate in organization A, but I joined and support organization B as well. \$\endgroup\$ – Scott Seidman Oct 16 '18 at 21:11
  • \$\begingroup\$ In fairness, I still try to nudge organization A to see things my way. \$\endgroup\$ – Scott Seidman Oct 16 '18 at 21:13
  • 1
    \$\begingroup\$ ... and of course, downvotes on this meta simply mean "I disagree", and not "you're wrong" \$\endgroup\$ – Scott Seidman Oct 16 '18 at 21:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .