12

There are (many) occasions where a questioner does not respond/acknowledge any answers to their question. Currently an answerer is prohibited from given any kind of recognition to their answer even after many days of non-response from the questioner. My feeling is that if someone asks a question then at least one of the responders should be able to take some kind of credit for devoting time and energy trying to be helpful.

  • 6
    I guess that most peoples (and ducks and marmots) will agree that this behavior is not nice. However, I'm wondering what your proposal is to change it. – marmot Feb 1 '18 at 20:15
  • 9
    Don't do good if you don't bear ingratitude (cit.). – CarLaTeX Feb 1 '18 at 20:36
  • 7
    You can help other users with this problem: go to tex.stackexchange.com/… (or whatever is currently the last page) and look for answers which are worth an upvote. I often find very interesting answers there. – user36296 Feb 1 '18 at 20:41
  • I think you may be able to start a bounty on the question and award it to yourself, though I haven't checked whether it's possible nor do I want to encourage the practice :-) – ShreevatsaR Feb 1 '18 at 22:04
  • 3
    @ShreevatsaR You cannot award a bounty to your own answer, see here: tex.stackexchange.com/help/bounty, the lines after How is a bounty awarded? – CarLaTeX Feb 1 '18 at 22:12
  • @CarLaTeX Ah makes sense, thanks! – ShreevatsaR Feb 1 '18 at 22:39
  • @ShreevatsaR: Apart from what CarLaTeX said: What would be the purpose of giving a bounty to yourself? You would lose rep and then gain it again ... (but it is impossible anyway) – user31729 Feb 2 '18 at 13:53
  • @ChristianHupfer The purpose of adding a bounty to the question is to draw attention to it (by having it show up in the Featured tab, etc.), so that more users can see the question and answer in context and upvote if appropriate. In fact if you expect to get at least N upvotes to your answer, then it still be a good “deal” to award a bounty of 10*N to another answer on the question. :-) Again, I don't encourage any such practice, was only thinking of possible solutions to the problem of no one reading your answer. – ShreevatsaR Feb 2 '18 at 16:23
  • 1
    I don't understand: there's no prohibition on answerers getting credit for their answers. Other users can give them credit. All they cannot do is accept an answer - only the questioner can do that. So I don't really see the problem: lots of answers get up-votes, even though the original questioner has disappeared. Bounties can also, as already mentioned, be used to attract attention to the question or reward other people's answers. Again, you don't have to be the OP to offer a bounty. – cfr Feb 3 '18 at 22:55
26

Remember that most of the 'value' of the site is meant to come not from dealing with the original poster but rather the wider user base. As such, a good answer should gain upvotes from several (or even many) people: that is the key to determining which answers are best. Answers being accepted by the original poster is certainly welcome, but it's far from the main determinant of answer ranking. As such, I'm not sure what one can say here beyond 'it happens'.

  • 3
    But if an answer does not get an acknowledgement of some kind from the questioner why should anyone else pay any attention or believe that the answer might actually be appropriate and/or correct? If there are several answers then let the system be and readers decide for themselves. If there is only one answer then perhaps questioners should be obliged to accept the answer or reject with reasons why. – Peter Wilson Feb 2 '18 at 19:04
  • I wanted to add [GOM] (Grumpy Old Man) to my comment but the system timed me out. – Peter Wilson Feb 2 '18 at 19:18
  • 2
    @PeterWilson The entire StackOverflow model is based on the idea of 'wisdom of crowds': lots of users look at answers and vote for those that work for them, which then have a higher ranking than other answers – Joseph Wright Feb 2 '18 at 19:20
  • I would like to discuss this offline. My email address is herries.press@earthlink.net. I should know yours but temporarily I can't find it. Any chance you can email me? – Peter Wilson Feb 2 '18 at 21:21
  • @PeterWilson The problem is that there is no way to “oblige” questioners on the site into doing anything. E.g. they may ask a question and never visit again, even to read the answers. Or they may read one answer, check whether it solves their problem (or give up) and never come back to the site to report. The only possible influence the site has is on repeat users (e.g. people who come back to ask another question, or post answers, etc.), but repeat users are not the majority of the “problem” questioners I imagine. – ShreevatsaR Feb 2 '18 at 22:17
  • 1
    @PeterWilson has a good point. I often search for information, not just about TeX, but other things that use StackExchange or a similar forum method. If none of the answers was ever accepted, then how am I (who knows much less) suppoesed to know that an answer works? There are many incorrect or misleading answers here and there. I suggest that if a question is good, but no answer is accepted after (say) a month, then users with sufficient privilege should be able to mark an answer as accepted. Odds are the OP simply doesn't care. – user139954 Feb 2 '18 at 23:22
  • @PeterWilson I'm joseph.wright@morningstar2.co.uk: note though that this is a network-wide idea and linked to the 'StackOverflow is not a forum' concept – Joseph Wright Feb 3 '18 at 17:50
  • Considering all the comments and the many times questioners are asked to "provide an MWE" or suchlike and they ignore the requests then there is no way some will follow procedures or requests. I'll just have to note any questioner I come across who does not acknowledge an answer and ignore any further requests from them. – Peter Wilson Feb 3 '18 at 20:03
  • @PeterWilson On my other network sites, such questions would get downvoted very quickly and then closed. That approach is likely reasonable on StackOverflow (professional programmers asking detailed questions), but here there's been a tendency to allow the 'benefit of the doubt' and leave open with comments. One may ask if it works ... – Joseph Wright Feb 3 '18 at 20:42
  • 3
    @RobtAll There are numerous examples of accepted answers suggesting poor solutions. If you rely on the green tick to assess quality, then more fool you. It is possibly marginally less foolish to rely on the number of upvotes, but still highly unreliable at best. – cfr Feb 3 '18 at 22:59
  • @JosephWright Speaking of other SE sites... I am not a programmer, so I often search for solutions to what would be routine info, using basic scripts. I can't tell you how many times I wanted to know how to do X, and the best questions and answers pertained for how to do X using Y on Z when P and not Q, However, I was unable to discern how to simply do X. – user139954 Feb 4 '18 at 0:27
  • 2
    So Zero or One - Vote answers are rubbish then... although they can be very tricky... – user31729 Feb 5 '18 at 12:43
12

This answer is more a response to Peter's comments on Joseph's answer.

The fact that a user accepts an answer or responds to it is neither necessary nor sufficient evidence for the validity of an answer anyway.

  1. Sometimes users accept an answer which is demonstrably worse than another competing answers. This can be for a variety of reasons, ranging from the speed at which users accept (accept the first answer that seems to work for them) to accepting an answer that although hackish, is at a level of code they can understand.

  2. The vote count on answers is often a reasonable proxy for the worth of the answer, so in the case of point 1 above, a competing answer may well have more votes than the accepted answer. This is especially true in our case, since we are quite diligent voters as a group, so good answers do tend to get upvotes.

  3. Good answers are very often accompanied with a complete MWE, even when the original question didn't include one. This fact alone makes it relatively simple for anyone to test whether a solution actually works or not (and non-working solutions rarely go unnoticed.)

11

This is the internet. Asking a question takes just a few seconds. We are a help desk, and we help free of charge.

You can end a phone call saying "Works, Thanks mate" and hang up without waiting for any response. Internet communication is not instant. You never know if the person asking the question is waiting for an answer, or asked a question, grabbed some luggage and goes on a holiday trip. Or the computer crashed 10 minutes after posting the question. Or the user posted on a smartphone while walking and fell into a pit ...

There is no guarantee to hear from the OP ever again, and there is no obligation to do so.

If the answer is good, it will attract upvotes by others.

You must log in to answer this question.

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