Everyone on this site has probably learnt something (perhaps with the exception of Jon Skeet) while here. That also means we've probably said something incredibly wrong, or didn't understand it enough to give a proper explanation.

I've been going through my 120~ answers I've made on Stack Overflow, deleting ones that I feel had no value, and now I'm trying to improve ones that I feel I can add better knowledge to.

For instance, I posted code that assigned global variables, but at the time, I thought it was right. I also just improved a question where I simply stated "That doesn't exist in spec A", but it does in spec B, so I added details about spec B and how it's used.

Generally I hope this improves the quality of both my answers, my ability to answer, and by effect, the site/community, which is really the crux of the problem I want to address.

We have a lot of bad knowledge out there.

We don't have any serious incentives in the game for most people to bother with it. We know the big rewards are in answering questions in the first ten minutes, but when someone comes from a web search to check us out, I've just given them bad information through my own fault, but nobody rewards me for improving the thing and thus a vicious cycle of programmers doing stupid things continues.

Can we reward people for fixing bad knowledge they left behind? Is this a non-issue?

  • I suspect it's a non-issue. If the answer is truly bad then it will be down-voted. Then you'll know what needs improving or deleting. – ChrisF Sep 20 '11 at 13:48
  • 4
    I imagine Jon Skeet has learned quite a bit on here. You don't get to 300K+ rep without asking/answering a few interesting questions along the way. – David Sep 20 '11 at 13:50
  • @ChrisF I see your point and it's something I've been debating as well. If the site's like a cross between a Wiki and other features, historically written information should be improving to better answers, regardless of how many votes it had. For instance, I've added information to an answer that was flatly missing from both answers, but I received an upvote on the original answer. – Incognito Sep 20 '11 at 13:52
  • 4
    @ChrisF: One issue with the approach of letting down-votes do the work is with unpopular questions. I don't have any examples right now, but I've seen questions with very low views and a single bad answer. Sometimes the answer even gets accepted, either because the person was told to increase their accept rate elsewhere or because, while filled with grossly incorrect information, the answer somehow provided a workaround for the asker's problem. I'm not sure if anything can be done in such a case, but it does leave bad information lying around for Google to index. – David Sep 20 '11 at 13:55
  • @David And that's what I'm after -- purge or improve the bad information. It's an up-hill battle to find consistently good information, and mistakes stick around for a long, long time. – Incognito Sep 20 '11 at 14:03

I don't think there is a way for Stack Exchange software to know if you are cleaning the wrong answer you gave, or you are simply correcting an answer that was already correct, but that needed improvements.
The only measure for bad answers is their score; still, for SE software an edit is just an edit, even if you edit a post that had a negative score.

There are two badges that are given to who edits posts that where inactive for 6 months. As far as I know, those badges are not given to who edit their own post, but it is still a badge given to who improves the old posts. Of course, that is not given to who deletes posts.
I think this is the closer we can get.

| improve this answer | |
  • It might be worthwhile to note, I sometimes answer questions over two years old, and get upvotes on those answers (over 10 in two cases alone). That may be useful here somewhere. – Incognito Sep 20 '11 at 14:08
  • Interesting detail which I myself encountered is that you cannot edit more than a certain amount of own old answers within 24 hours. I believe the limit is max 5 answers which are over 6 months old (not that I only post bad answers, I just keep improving them :) ). – BalusC Sep 20 '11 at 14:10
  • @Incognito When you edit an answer you gave, the question containing it is bumped in the front page, and users can vote your answer at their will. If you are speaking of an "automatic" vote, I don't think that should be granted. – kiamlaluno Sep 20 '11 at 14:16

You must log in to answer this question.

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