3

When searching "Tabs vs Spaces" on Google, the very first result is "Tabs versus spaces—what is the proper indentation character for everything, in every situation, ever?"

Google search for "tabs vs spaces"

Despite being closed and locked as non-constructive back in 2011, the question has been viewed over 418,000 times, making it one of the site's most popular questions. (The post was presumably closed because Tabs vs Spaces is such a controversial issue.)

So if that's the case, why does a subjective question about Tabs vs Spaces have an accepted answer?

Apparently, spaces is the *correct* answer

There are several reasons this question shouldn't have an accepted answer:

  • It implies that Software Engineering endorses one coding style as the correct answer
  • Most novice programmers will likely follow the first answer they see
  • Community wiki questions hide the author, giving a false impression that the site itself takes control of the post's answer
  • The accepted answer has half the votes of the top voted answer (+129 vs +249)
  • The post was locked for not having a definitive answer, but the lack of voting implies that the site considers spaces as the single definitive answer
  • While active users of the site know the question doesn't represent the site, the thousands of new users will likely ignore the locked prompt and go straight to the answer.

The least we can do is avoid implying that this question has a correct answer.

  • 2
    What, exactly, do you propose we do? The question has over 400k views - deleting it is out of the question since deletion would probably break many links. It's not possible to unaccept an answer, unless the original question asker does so. I (or any moderator) could undo the Community Wiki aspect, but would that actually solve any problems? – Thomas Owens Oct 26 '17 at 16:52
  • @ThomasOwens The OP is still active on SE. Is it possible for a question author to unaccept answers on a locked post? – Stevoisiak Oct 26 '17 at 17:07
  • 3
    this is what should be done to stop confusing visitors: feature request at our meta / same at MSE "about 15 millions views at Stack Overflow are on inappropriate (historically locked) questions... similar analysis at Software Engineering shows about 5M views at historically locked questions over there..." Quite a pity that SE team doesn't give a sh!t – gnat Oct 26 '17 at 18:18
  • @StevenVascellaro, I see very little evidence that the OP is still active on Software Engineering, so I think it is unlikely that the OP will take any action. – Bart van Ingen Schenau Oct 28 '17 at 17:13
  • Along with vi/emacs, the tabs/spaces debate is a dumpster fire of a religious war. Best just to ignore it. – user22815 Nov 11 '17 at 9:44
  • 3
    "Tabs vs spaces" has an accepted answer because the OP accepted an answer. Stack Exchange's rules allow the OP to accept any answer he wants, for any reason he wants. – Robert Harvey Dec 6 '17 at 20:46
  • @StevenVascellaro Don't see how following the first answer is bad in this case: using spaces is exactly what novice programmer should do as he won't be able to use tabs correctly anyway. – StaceyGirl Dec 24 '17 at 2:04
4

First, you surely noted that your question was already asked in a comment in 2011, and that comment was already answered in the following comment:

Remember that an accepted answer is not an answer that is universally accepted from all the users; it's the answer that is accepted from the OP, who chooses the answer that helped him more. Only the OP can know which answer helped him more than other answers; even in the case of two questions that basically say the same thing, there is still an answer that can help the OP more than the other. – kiamlaluno Sep 29 '11 at 13:01

Second, I think just unaccepting the question would not solve the problem either, this would bring the answer "Tabs" to the top, which is not better than seeing "Spaces" as the top answer. Quite the opposite, currently the fact "the accepted answer has half the votes of the top voted answer" works IMHO to some degree as a warning sign not to trust that accepted answer blindly.

So, my best recommendation here is:

Learn to live with the fact that the SE sites are not perfect, and not every question's accepted answer is the correct, best, finest answer for everyone, everywhere and everytime.

Besides that, I think you are overstating the "problem" for real world development here a little bit. If some novice programmer will be so dumb and follow blindly the first answer they see on this site, given by some strangers from the internet, (and ignore the fact the second best answer got twice as many upvotes, though saying 100% the opposite), then we cannot help them.

You must log in to answer this question.

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