3
\$\begingroup\$

This issue has popped up before, and again this morning, and I think it deserves some attention.

What happens is this: Alice asks a question. 5 minutes and 2 views later Bob answers. Another 2 minutes and 1 view later Alice reads the answer, problem solved, so she upvotes and accepts the answer.
Carl who also had a great answer feels he didn't get a proper chance and leaves without answering at all. The thread may die there after 3 views.

Rep is the payment we get for answers (and questions) and we all appreciate recognition, so I can understand Carl's reaction. I also told the asker of a question that he accepted too quickly a few times, even when it was my own answer which was accepted, just because I also feel it isn't fair to others.

Comments say that it's always possible to accept another answer, but this rarely happens. I guess the asker doesn't come back to read other answers at all, why else would she accept so quickly.

So that's the situation. Can and should we do something about it? I think we should. Federico suggested a minimum of views before an answer can be accepted. I think this is a good idea. A minimum time (like a day?) between asking and accepting could also be a solution.
What do others think?

edit
Since I posted this I've had the honor to comment to OPs about this a zillion times :-(. My advice is now to wait at least a day or until 3 or 4 answers are posted. The day is to allow the question to go around the world. If Alice in Europe post at 9 o'clock in the morning local time, and Bob, also in Europe, answers at 9:01 :-) then the answer may be accepted before Carl in the US even gets out of bed. That's what I mean by "letting the question go around the world".

\$\endgroup\$
  • 2
    \$\begingroup\$ how many will not be accepted when someone gets a perfect answer and decides they are happy. Wait, they cannot accept, well, no harm to them. On the way they go. Now no one gets accepted. \$\endgroup\$ – Kortuk Jul 12 '11 at 6:56
  • 1
    \$\begingroup\$ @Kortuk - I see your point. But are we interested in one-time visitors who just want an answer to this one problem they have, or people who are really interested in multiple points of view, and want to actively support EE? \$\endgroup\$ – stevenvh Jul 12 '11 at 7:01
  • 1
    \$\begingroup\$ I am in interested in someone whom wrote an acceptable answer getting the credit they deserve instead of no one receiving credit. The community member whom stretched their neck out and wrote something at the drop of the hat to help probably deserves their credit. Maybe we as a community should pay more attention to these types of events. \$\endgroup\$ – Kortuk Jul 12 '11 at 12:12
  • 2
    \$\begingroup\$ if you see a "late but great" answer please come post it to chat, there are at least 5-10 people that will see it and often support it. \$\endgroup\$ – Kortuk Jul 12 '11 at 12:13
  • 4
    \$\begingroup\$ Looking at other SE sites where I have 0% accept rate because 100% of the answers I get are unhelpful, I'm happy that electronics.SE has this "problem" of people answering questions too well, too quickly. :) \$\endgroup\$ – endolith Jul 14 '11 at 21:01
  • 1
    \$\begingroup\$ If disabling accept for a predefined period/number of views is not acceptable, why not pop up a warning box giving the explanation several of us leave in a related comments. \$\endgroup\$ – jippie Aug 18 '12 at 13:48
12
\$\begingroup\$

I would even be happy with just adding a box that says something like:

Wait, are you sure you want to accept this answer, there has only been 10 views? You might get a better answer later.

I think this would solve several of the issues that people have brought up.

\$\endgroup\$
  • \$\begingroup\$ Isn't this the same as my answer? :) \$\endgroup\$ – endolith Jul 12 '11 at 19:51
  • \$\begingroup\$ @endolith look at the comment chain on jeff's answer \$\endgroup\$ – Kellenjb Jul 12 '11 at 19:52
  • \$\begingroup\$ @endolith, I think both answers are good, but they seem different to me. \$\endgroup\$ – Kortuk Jul 13 '11 at 14:23
  • 2
    \$\begingroup\$ @endolith: You did mention it, but he said it better and clearer. \$\endgroup\$ – Olin Lathrop Apr 10 '12 at 12:19
8
\$\begingroup\$

Yes. If the problem's solved, it's solved. Maybe the accepted answer is wrong and a better one will be posted later, but maybe the asker tested it and confirmed 100% that the answer is correct, so why shouldn't they be able to accept it?

Forcefully preventing the asker from accepting an answer based on an arbitrary time or view number is dumb. A dialog reminding them there might be better answers and encouraging them to give it a little more time is sufficient.

\$\endgroup\$
  • 1
    \$\begingroup\$ Yeah, thinking about it the popup dialog is a good idea. Kellenjb also mentioned it in a comment. I think this could be a good compromise. \$\endgroup\$ – stevenvh Jul 12 '11 at 15:57
  • \$\begingroup\$ Problem being solved is not the only issue. That makes the answer correct, but not necessarily the best. Another answer could suggest the same solution but provide more background or insight. However, I'm liking the popup idea. It has only been 17 minutes since you asked the question. Others may not post answers after you accept one, even though they could provide additional, more complete, or better information. Do you really want to accept this answer now? YES - CANCEL. \$\endgroup\$ – Olin Lathrop Jul 12 '11 at 18:01
  • \$\begingroup\$ @Olin - Nah, no need for the "OK" button, just "Cancel" :-) \$\endgroup\$ – stevenvh Jul 12 '11 at 19:36
  • 1
    \$\begingroup\$ Seems we all can live with the idea of the popup dialog. Is this something that can be implemented? \$\endgroup\$ – stevenvh Jul 12 '11 at 19:39
  • \$\begingroup\$ @Olin: If the correct answer has been posted, but it could be improved, then you should edit it to improve it. Posting a duplicate answer that says the same thing in a different way, purely so that you get reputation points is not the way the site is supposed to work. \$\endgroup\$ – endolith Jul 12 '11 at 19:44
  • \$\begingroup\$ @endolith: I didn't mean saying the same thing in a different way, but containing significant more relevant information, background, why it works instead of just what to do, mention of more alternatives with their pros and cons, etc, even if the result of the analysis is the same as the previous answer. \$\endgroup\$ – Olin Lathrop Jul 12 '11 at 23:32
  • 2
    \$\begingroup\$ I agree with Endolith - If you have something you're happy with, then just click it! People can always provide new answers and comments - also you can just change your accepted answer at any time \$\endgroup\$ – Jim Jul 14 '11 at 8:35
  • 1
    \$\begingroup\$ I guess it depends on the particular answer / question, some things will just have a definite answer and there's little need to say more \$\endgroup\$ – Jim Jul 14 '11 at 8:38
4
\$\begingroup\$

Per

Discourage questions being marked as answered within an hour or so of being posted

We prevent accepting for 15 minutes after asking.

I don't think I would support basing this on views at all.

\$\endgroup\$
  • 1
    \$\begingroup\$ Why not views? Maybe some argument I may have overlooked? And 15 minutes is not that long. I was more thinking about 24 hours or so. \$\endgroup\$ – stevenvh Jul 12 '11 at 8:59
  • \$\begingroup\$ For this question I have one answer after 24 hours, and 61 views. Still hesitate to accept the answer, because I would like to see other points of view. \$\endgroup\$ – Federico Russo Jul 12 '11 at 9:21
  • 1
    \$\begingroup\$ @Jeff - 15 minutes is a long time on Stack Overflow, but very quick on low-traffic sites like ours. Would you consider adjusting this number based on site traffic to something closer to 30 minutes or an hour? \$\endgroup\$ – Kevin Vermeer Jul 12 '11 at 12:15
  • 1
    \$\begingroup\$ Also, I'm as puzzled as Steven is as to why you wouldn't support basing this on views. \$\endgroup\$ – Kevin Vermeer Jul 12 '11 at 12:16
  • 2
    \$\begingroup\$ I would even be happy with just adding a box that says something like "Wait, are you sure you want to accept this answer, there has only been 10 views? You might get a better answer later." \$\endgroup\$ – Kellenjb Jul 12 '11 at 12:36
  • 2
    \$\begingroup\$ Who wants to have to wait until many people view their question before they can accept a great answer. \$\endgroup\$ – Kortuk Jul 12 '11 at 13:57
  • 1
    \$\begingroup\$ @Kortuk: But how do you know it's a great answer until you know the alternatives and people have had a chance to comment? Comments are like peer review. Since the OP is the one asking, they are not in a good position to judge the greatness of a stand-alone answer. Even in cases where a answer demonstrably solves the OP's problem, another answer could provide better background or understanding. \$\endgroup\$ – Olin Lathrop Jul 12 '11 at 15:00
  • 1
    \$\begingroup\$ @Kellenjb, Can post that suggestion as an answer? (Having a confirmation window if the time taken to select a "correct" answer was considered short) \$\endgroup\$ – Jon L Jul 12 '11 at 19:14
  • \$\begingroup\$ @Jon done... and some filler for the comment length requirement \$\endgroup\$ – Kellenjb Jul 12 '11 at 19:42
3
\$\begingroup\$

I'm not convinced there's such a problem here, it may just be that we're over-thinking things. Atwood had a great post on Coding Horror, about the perils of meta-discussions, and of course that's what this Meta site is for, but it seems like a lot of time and effort is going into this discussion, effort that could be going into the main content site.

It seems like this is only really a problem for rep hunters, and a possibly hitherto unidentified sub-group, the green check-mark hunters. What difference does it make, if Alice gets a useful answer five minutes after hitting 'post'? Does the SE exist to help people solve problems, or does it exist to stroke the egos of the regulars? I would hope it's the former; personally, I got involved here b/c of repeatedly getting burned by search results turning up nothing but "yeah, here's your question, now PAY UP to see the answer". And I may get uninvolved here at some point when the popularity-contest aspect of it becomes too overt.

So, if someone feels they got they answer they need, why should they be obligated to wait around? I think those of us that have spent more time with the SE will tend to let questions ride a while, but casual passers may not realize the etiquette that has evolved. If their visit is a one-off event, so be it, at least they helped seed some content. If they liked the experience (and arguably, quick deployment of an 'accept' should be taken as indication of a satisfied customer), they are more likely to come back, and if they do, they will soon enough learn to let the question marinate in the attention of the community a while longer.

My point is, people will come here with questions because the have problems they want to solve, and need answers - but I'm not so sure they will keep coming back if the focus is changed to force them to be judges in a popularity contest.

\$\endgroup\$
  • 1
    \$\begingroup\$ You are looking at this only from the point of view of someone wanting to get answers. Nobody does anything truly for free. Around here the "pay" is rep. The numerical rep gives you some priveledges, which don't matter much after the first few 1000 points. However, high rep gives you less measurable but still very real intellectual rep and perceived quality. That stuff is important, so the score keeping on this site is important, at least to some of us who write a lot of answers and do consulting for a living. \$\endgroup\$ – Olin Lathrop Apr 10 '12 at 12:33
  • \$\begingroup\$ maybe so, but without the questions, what would be the point? and I don't see a problem being an advocate for the askers here, b/c the very nature of the meta site makes it of more interest to the answerers than to the askers, so the latter's interest is likely under-represented here. \$\endgroup\$ – JustJeff Apr 11 '12 at 2:32
  • \$\begingroup\$ I sometimes think it would be entertaining to know what the dynamic would have been if rep points were only visible on profile views. \$\endgroup\$ – JustJeff Apr 11 '12 at 3:42
  • \$\begingroup\$ You seem to have some objection to people that spend a lot of time answering questions getting something out of it, even if it costs the askers nothing. This makes no sense, because even from purely the asker's point of view this is in their interest since it encourages more and better answers. \$\endgroup\$ – Olin Lathrop Apr 11 '12 at 14:59
  • 2
    \$\begingroup\$ I just think that there are two broad categories of answerers - those who do it mainly for the rep points, and those who would do it anyway. That the former wouldn't understand the latter is not entirely surprising. \$\endgroup\$ – JustJeff Apr 12 '12 at 3:14
  • \$\begingroup\$ But other than that you're on some vandetta against people getting recognition for their work, why do you care? The point is to get good answers, and the recognition a good rep system provides is one way to encourage that. Since this comes for free to those looking for answers, why are you so against it? As long as someone provides good answers, you shouldn't care why they do it. \$\endgroup\$ – Olin Lathrop Apr 12 '12 at 18:02
  • \$\begingroup\$ Also, despite what people may claim, nobody truly does anything for free. They may get personal satifaction, a warm feeling, or something else intangeable out of it, but everyone is in it for something. If your goal is getting the best answers, then you should want to enhance all the little ways people feel rewarded for doing so, including the intellectual recognition that rep points here provide. I don't understand why you feel so threatened by others earning rep. It's not coming at any cost to you. \$\endgroup\$ – Olin Lathrop Apr 12 '12 at 18:07
  • 2
    \$\begingroup\$ "vendetta"? I think not. There's no need to cast me as "against" anything, when I plainly said what I was for, which is people getting answers, w/o a lot of roadblocks getting in the way. And "threatened"? Why would I feel threatened? \$\endgroup\$ – JustJeff Apr 13 '12 at 23:52
2
\$\begingroup\$

Similar issues have been discussed on the main meta.

I don't see any reason to delay accepting answers. We want people to indicate when something solved their problem. People not accepting answers at all seems to be a bigger problem them accepting them "too quickly".

I suppose some people probably are discouraged from answering if one has already been accepted, but they shouldn't be. If Carl provides an answer that is noticeably more thorough or useful than Bob's, he will come out ahead in rep over the long term, regardless of who's was first or was accepted.

Conversely, if Carl has what amounts to a "ditto" answer, discouraging him from posting it is a fine thing to do.

\$\endgroup\$
-4
\$\begingroup\$

I think using both a time limit and number of views, whichever occurs first, makes sense. The point is the OP can't know what other answers are possible until enough people have had enough time to write their answers. The number of views is probably the best single measure, since that says how many people have had a chance to answer the question. The time limit is more of a safety in that you can't hang the OP out there forever. At some point you have to say enough is enough, even if few people viewed the question.

I do think that the current 15 minute time limit is way too short. If you want to base it on time limit alone, then I think 24 hours makes sense. Many people may check in once a day, so that leaves time for one normal cycle.

I just took a quick look at recent questions that were 12 hours or older, and it seems like 50 views might be a reasonable threshold.

Number of answers could also be relevant. Once you have at least 3 or maybe 4 choices, I'd say it's reasonable to pick one.

\$\endgroup\$
  • 5
    \$\begingroup\$ I don't support the minimum answers idea. Some questions get only one answer, so you won't be able to accept at all. I vote for 50 views. \$\endgroup\$ – stevenvh Jul 12 '11 at 15:03
  • \$\begingroup\$ @stevenvh: It was just a thought. In any case, I only meant it to be a sufficient requirement, not a necessary one. In other words, even if it hasn't been very long and few people viewed the question, it's still OK to pick one if there are 4 to pick from. \$\endgroup\$ – Olin Lathrop Jul 12 '11 at 17:55
  • \$\begingroup\$ Yes, I agree that sufficient answers can override the time/views restriction. What do you think of endolith's popup dialog, Olin? \$\endgroup\$ – stevenvh Jul 12 '11 at 17:58

You must log in to answer this question.

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