-2
$\begingroup$

At the moment there is a competition for being the first answer because the first answer tends to earn the most votes. As a consequence some people tend to only skim the question or even only half of it before they start writing an "answer". Again as a consequence, the overall quality of answers drops since once a question has an "answer" it ends up getting less views reducing the chance of getting an accurate and to the point answer.

I have been thinking about this for more than a year and yesterday I think a possible solution to the dilemma occurred to me which I hereby would like to propose:

How about we introduce a 30 minute threshold of time after a question has been posted during which the thread is editable but posted answers will not appear? It means you can write your answer, post it and edit it but no one will see it until 30 minutes are up.

As a consequence I think the competing for "time" would shift to being a competition for "accuracy and quality".

Addendum: If anyone else has any suggestions on how to cut down on suboptimal answers please contribute. I think it is a big problem. Answering should not be a race for time but a strive for quality and excellence.

$\endgroup$
27
  • 16
    $\begingroup$ (a) There are a lot of questions which can be reasonably answered in less than 30 minutes of its posting (b) Such a change almost surely will involve a change to the entire SE backend, and I am not optimistic that users of other SEs will share your perspective. $\endgroup$ – Willie Wong May 29 '13 at 9:29
  • 2
    $\begingroup$ @WillieWong I know, I thought about (a). My thoughts on it are that for questions like e.g. a reference request or a yes or no question or similar that can be answered in a minute it would not be a problem that the answers would appear after 30 minutes. Some of these are already answered only half an hour or more later. The way I see it is that there is no mathematical question that is so pressing that it cannot wait for 30 minutes. $\endgroup$ – Rudy the Reindeer May 29 '13 at 9:33
  • 1
    $\begingroup$ As for the additional work on the backend: I can write code and I have built websites. This is not such a big thing. The other SE sites would not even have to know that we have this feature. $\endgroup$ – Rudy the Reindeer May 29 '13 at 9:35
  • 5
    $\begingroup$ @MattN. Client-side restrictions can be easily circumvented. Server-side modifications predicate on StackExchange letting your modify the code that is on their server. Unless you are a SE employee/intern, I cannot see that happening. $\endgroup$ – Willie Wong May 29 '13 at 10:03
  • 2
    $\begingroup$ A related question. $\endgroup$ – J. M. isn't a mathematician May 29 '13 at 10:16
  • 6
    $\begingroup$ Also related: over on Meta.SO this is called the Fastest Gun In The West problem. There have been extensive discussions there. See, e.g. this old blog entry, and this tag. $\endgroup$ – Willie Wong May 29 '13 at 10:19
  • 3
    $\begingroup$ @WillieWong Alright. But I do want to "discourage the quick and dirty answer". Because on math.SE, quick and dirty mostly translates to "beside the point". $\endgroup$ – Rudy the Reindeer May 29 '13 at 10:27
  • 13
    $\begingroup$ With the change (and even before it), I think the third point in Tom Oldfield's answer below has to be considered, and not just for rushed answers, but even good answers. It happens numerous times that a question is posted which has an almost canonical answer. If everyone is blind as to whether an answer has been posted for 30 mins, it would be easy for virtually the same answer to show up 30 times after this blackout period. This is likely more problematic than the problem it it supposed to solve. $\endgroup$ – user642796 May 29 '13 at 10:35
  • 5
    $\begingroup$ Wouldn't it be an enormous waste of time and effort of MSE users to find out after 30 minutes that several of our top users have spent time on answering something trivial, just to find out that the question had, in fact, already been answered by several other people and the answers had been hidden? $\endgroup$ – Martin Sleziak May 29 '13 at 12:15
  • 2
    $\begingroup$ See my modest proposal of 7 Nov 2011, posted as a comment at meta.math.stackexchange.com/questions/3183/… (and see, too, the reaction to it). $\endgroup$ – Gerry Myerson May 29 '13 at 12:18
  • 5
    $\begingroup$ Given that the scenario described by Arthur Fischer happens even now - essentially identical answers appearing in the span of minute or two (= the time it takes to type a canonical answer). This problem would become exponentially worse. Just downvote the answers that missed the point. $\endgroup$ – Jyrki Lahtonen May 29 '13 at 12:28
  • 2
    $\begingroup$ @MartinSleziak Actually the waste of time happens on my side if I spend 30 minutes or more to write a good question as clearly as possible just to get blah-blah-blah as an answer within 5 minutes. Also, likely no good answers to appear after one is there already since people tend to look at threads with no answers. $\endgroup$ – Rudy the Reindeer May 29 '13 at 12:29
  • 4
    $\begingroup$ If the answer missed something, or is otherwise bad, yes, you can downvote. You can downvote even if the answer was correct. You can upvote/downvote for whatever reason you see fit! $\endgroup$ – Jyrki Lahtonen May 29 '13 at 12:32
  • 3
    $\begingroup$ @JyrkiLahtonen I'm not sure I will start downvoting again. I tried it and it had no effect on the answers. Even completely wrong ones remain. The only effect downvoting seems to have is to drag down the general ambience here. $\endgroup$ – Rudy the Reindeer May 29 '13 at 12:37
  • 3
    $\begingroup$ For questions with canonical answers at any given time there are likely tens of users logged in with knowledge of this answer who could completely type it up and submit it within the blackout period. If each is unaware that others are working on it, each is much more likely to expend the effort typing it up and submitting it. The net effect will be certain questions flooded with answers, which, IMHO, is not an optimal situation. Furthermore, as more users are typing canonical answers, non-standard questions may actually see less answers; again not an optimal situation. $\endgroup$ – user642796 May 29 '13 at 13:16
16
$\begingroup$

I think that there are a couple of things wrong with this idea, although I appreciate the sentiment that you would like to improve the overall quality of answers.

  • It would be hard to tell from the active queue whether or not a particular question was open for answering. This would be frustrating if there was a question that readers were interested in, and then find themselves unable to answer once clicking on it. Things like this could lead to a greater likelihood of questions being ignored. If whether or not a question is "open" was open was displayed on the queues, this could well lead to questions being ignored to some extent until they become open, again a bad thing.

  • This could cause a lot of duplicated effort. If two or more people work on very detailed, almost identical answers for a significant time period during the $30$ minutes where answers can't be posted, questions may end up with many similar answers, annoying people reading the question and the people who answered it.

  • Users who usually rush to write answers may simply write answers in a new tab, to post immediately after the deadline, leading to a flood of answers on each question after the $30$ minutes. The answers may be less rushed, but there would be a lot of them, making it difficult to go through them all fairly. Questions which attract many, very similar answers already (for example because the required answer is quite short and relatively "easy") may be affected particularly strongly.

  • Personally, I like to know how long a question has been posted, and see how many answers it has. If there are many answers in a short time period, I am likely to suspect that the question has not much more to it than what appears in the title, and move on unless it seems particularly interesting. If it has been up for ten or more minutes, and has none or very few answers, than I am more likely to think that the question is a little deeper and read it, even if the title gives the impression that the question is a little "simple". I think that the number of answers can be just as good an indicator of a good (or interesting) question as the number of upvotes, and it would be a shame not to have this information on questions in their first half hour.

$\endgroup$
2
  • 2
    $\begingroup$ I wasn't specific enough: the edit field should be available at all times so that your point one is not a problem. You "post" your answer but it won't appear and remains editable until the 30 minutes are up. Let me edit my suggestion. $\endgroup$ – Rudy the Reindeer May 29 '13 at 10:20
  • 1
    $\begingroup$ @MattN. Fair amendment, but it may be the case that people still would prefer not to answer questions in the "blackout" period for some of the other reasons in the post so I think I'll leave the point up there. $\endgroup$ – Tom Oldfield May 29 '13 at 21:39

You must log in to answer this question.

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