Many questions on Programmers.SE are complex, behavior related and subjective ("My team lead is a psycho", "How to best teach coding", "Meetings in agile" etc). The nuances and depth of the questions and answers is what should define this site.

Some of these complex questions receives a bunch of easily digested, clever answers which gets upvoted quickly and then marked as accepted just as fast. That is fine on Stackoverflow, but programmers.SE are a completely different beast which deserves long and complex answers. I think there should be some way of actively encourage that. It's about increasing the quality of answers really.

A suggestion I thought about:

Don't allow questions to be marked answered until after a certain "grace time" (3-4 days) has passed.

Rationale: Questions left without an accepted answer for a number of days seems to shift. The easy/quick/clever answers loses ground to more thoughtful answers which bubbles to the top. That is a good thing. I assume it's because it takes time to digest the complexity and the hidden brilliance in some of the answers. Adding a grace time can perhaps help encourage that.

Other ideas?


Extending the time limit for accepting answers addresses the wrong facet of the problem. The central issue is not a lack of thoughtful answers, but an overabundance of thoughtless answers, and a 3-day waiting period will do precisely nothing to hold back the tide of one-liners and hyper-generous Civic Duty-seeking voters.

I suggest swift and merciless downvoting of thoughtless answers, and moderator-flagging of truly useless (joke or retort) answers. That's how this site, and every SE site, is designed to be used.

It also wouldn't hurt to point these members to the FAQ, politely explaining that the question guidelines also apply broadly to answers. In other words, explain your answer, back it up, and mind your tone.

This may be hard for some people to believe, but on other SE sites, we've actually had fairly mindless poll questions saved by early, thoughtful, detailed answers. It's a matter of instilling those values in the community, and the way to do that is not with arbitrary and frustrating technical limitations, but with behaviour - positive and negative reinforcement - and ongoing discussion and education.

I don't know about you guys, but my reputation on Programmers.SE means about as much to me as my Meta.SO rep, which is to say, not much. Just downvote, guys; stop worrying about the -1 rep hit you'll take. It doesn't matter if 10 people upvote a stupid answer when there are 30 more waiting to downvote it, and eventually after seeing the negative scores, people will start to take the hint that unconstructive answers are frowned upon.

(And yes, I am aware of the unwritten law that any meta answer encouraging downvoting must be downvoted several times, so go ahead and downvote it - just don't forget to downvote on the parent site, too.)

P.S. Increasing the weight of downvotes probably wouldn't hurt either. It's not appropriate for Stack Overflow, but it makes a certain amount of sense for subjectivity-oriented sites because people are more likely to post controversial or rep-whoring answers.

  • 1
    I like the idea of downvotes being weighted more. Would you consider removing the rep penalty for downvoting too? – Michael K Dec 9 '10 at 17:33
  • 1
    @Michael: Whether I would consider it isn't really important; the question is whether or not Jeff would consider it, and having participated in several downvoting-related discussions on Meta.SO, I can tell you without hesitation that he would never consider it. The weight issue is more important anyway, because the disproportionate weighting rewards controversial answers. – Aaronaught Dec 9 '10 at 18:51
  • @Aaronaught: Would it be possible to try the time limit as an experiment and see how it goes? Inspect and adapt? I still cling to the idea that great answers needs time to get upvoted. – Martin Wickman Dec 9 '10 at 20:52
  • @Martin: You'll have to ask the development team, but I doubt it. You have yet to even show any examples of how time-to-accept can negatively affect answer/vote quality, much less show any correlation, which even then is several steps away from proving a causal relationship. On the other hand, you can try downvoting right now - no changes required! – Aaronaught Dec 9 '10 at 21:33
  • @Aaronaught: Still think it could be worth a try how it plays out. It's not the end of the world. Then we'll evaluate it. – Martin Wickman Dec 10 '10 at 6:45
  • @Martin: Making random changes on a live site just to see what happens is not a good development practice and I'm pretty sure the team will agree with me on that. Show us that it's actually likely to effect positive change and maybe it'll get implemented. – Aaronaught Dec 10 '10 at 16:39
  • @Aaronaught: I never mentioned how it should be implemented. It is just a suggestion! Don't even begin lecturing me about "development practices". That's got nothing to do with this, and you know it. Besides, trying stuff out is how it's normally done on the web, just look at Google. I think the team will agree with me on that. This all smells NIH. – Martin Wickman Dec 10 '10 at 17:32
  • 2
    @Martin: See Politician's Syllogism. – Aaronaught Dec 10 '10 at 20:31
  • 1
    Yes @Martin, and I discussed in detail why it's not a good suggestion. I don't know what your problem is; you seem to think that they should go and implement for no other reason than because you suggested it. That's not the way it works; you have to convince people that it's even worth trying. – Aaronaught Dec 11 '10 at 0:52
  • 1
    @Aaronaught So far only your (rather conservative) voice has been heard regarding this suggestion, and that hardly constitutes a consensus agreement from the Powers That Be. So we don't know what they think and I'm not that invested in this to bother to convince anyone more than what I said in my original question. – Martin Wickman Dec 11 '10 at 9:11
  • 1
    @Martin: If you're not even willing to put in a modicum of effort to find some examples to support your case then your voice isn't going to get heard. Yeah, I don't speak for the team, but I've spent over a year on Meta.SO, so trust me on this one. Probably the only reason this isn't [status-declined] already is that you worded it as a discussion question, even though you tagged and are handling it as a feature request. – Aaronaught Dec 11 '10 at 16:00
  • 3
    @Josh: No, it doesn't make sense. Even if the comic is directly relevant, it's not an answer. It's not factual evidence nor is it reflective in any way of the author's actual experience. It's nothing more than a witticism, appropriating somebody else's content and claiming partial ownership (not of the intellectual property itself, but in the sense of "Look, I was the first person to make the connection!"). Q&A sites are about providing helpful answers, not amusing ones. Comics and witticisms belong in the comments, not answers. – Aaronaught Dec 13 '10 at 18:37
  • 3
    Bottom line: Far too many people essentially treat this site as a glorified discussion forum, spewing out the first thought that penetrates their skull, no matter how irrelevant, unverifiable, or frivolous it is. Massively-upvoted answers that contain nothing but a comic strip drawn by somebody else are merely one of the most glaring examples of that groupthink. – Aaronaught Dec 13 '10 at 18:43
  • 1
    @Aaronaught: As apposed to massively upvoted missives that boil down to nothing substantial? – Josh K Dec 13 '10 at 18:49
  • 2
    That's an impressive trainwreck of a Q&A thread, @Josh. All I can say is that I didn't vote for any of it... – Aaronaught Dec 13 '10 at 18:56

Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction.
-Albert Einstein

I dispute the notion that easily digestible, clever one-liners aren't thoughtful answers. Oftentimes they take more thought than longer answers.

  • 1
    Examples would be nice. I'm pretty sure that Einstein didn't have this in mind when he made that quotation. – Aaronaught Dec 11 '10 at 15:53
  • @Aaronaught - Don't have any. I'm just making the case that long and complex != good and thoughtful. And yes, I do believe that Einstein would have agreed that an easily understood, clever answer would be preferable to and more work than a long and complex answer. – Jason Baker Dec 12 '10 at 2:56
  • Einstein worked in a patent office. He would have demanded evidence and so should everybody here. There's a world of difference between conciseness and superficiality. – Aaronaught Dec 12 '10 at 3:02
  • 1
    P.S. Einstein is also quoted as saying Everything should be made as simple as possible, but not simpler. This community would be wise to remember the last three words. – Aaronaught Dec 12 '10 at 3:03
  • 2
    This reminds me of Blaise Pascal's "I would have written a shorter letter, but I did not have the time." By which I don't mean to back up Jason as much as @Aaronaught. Take your time to compose an answer that is as comprehensive as possible, and then take your time to make it as concise as possible. I am pretty certain that posting a half-assed quip within seconds of a question being asked and leaving it at that is not what Einstein or Pascal had in mind. If anything, such quips should be left as comments. – RegDwight Dec 14 '10 at 13:30

You must log in to answer this question.

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