What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

I asked the following question on WebApps SE:

Can I view Google+ check-ins on Google Maps?

With Google Latitude closing down next month, this invalidates the correct answer to the question.

Should I:

  • Unmark the answer as correct and edit my question
  • Open a new question stating "...without Google Latitude"

The first solution means the original answerer will lose 15 points, even though, at the time, they were correct.

The second solution means the original question will still be available to the public, when its no longer correct / helpful.


I believe the first solution would be best, but I think the original answerer deserves reward for their help.

Even if I remove the "answer mark" the user could start getting negative votes:

"-1 Google Latitude is no longer exists"

share|improve this question
    
I would hope people don't start down-voting that way just because Latitude no longer exists. The answerer obviously couldn't have predicted that at the time they posted their answer. Now, if two months from now someone posts an answer about Latitude, that would be a different story. –  Aaron Bertrand Jul 10 '13 at 14:12
2  
I'm confused - everyone's talking about 'the answerer' but unless I'm losing my mind you are both the asker and the answerer of that question? You don't get rep for accepting your own answers so you won't lose any by unaccepting it... –  dsg Jul 10 '13 at 14:49
2  
@user216011 Ha to be honest I hadn't even realised I was the answerer to that question. My question is more generic than this particular, rare, scenario though. So my point still stands, but well spotted. –  Curt Jul 10 '13 at 15:18

3 Answers 3

up vote 5 down vote accepted

Quite honestly, if the answer helped you and was correct at the time, and there isn't an alternative answer that is now more correct, I'd leave it be. We can't be perpetual janitors checking for any answers that are now obsolete because something outside of the site or our control has changed. As an example, what are we going to do with over 3,000 Google Reader questions? Wipe them clean off of the site? Forget about someone losing 15 measly rep, that could lead to people changing rep tax brackets.

In the SQL Server world, this kind of thing happens all the time. A mock question from 2009:

  • Q: "What is the best way to do x?"
  • A: "Well, since SQL Server doesn't support foo, the most efficient way is..."

Two versions later, and SQL Server now supports foo. Do we go back to a 4-year old question and change the accepted answer because there is a better answer now? Probably not, and it's actually a pretty dicey problem, since some readers who come across the question will have the latest version and some will still be using the same version with the same limitations.

I guess I would suggest adding a comment to the answer (or even suggesting an edit to the answer) with a disclaimer that Latitude is dead. This will give the answerer the first opportunity to maintain their accepted status by updating the answer with a new alternative (if one exists).

share|improve this answer

Generally it's a good thing to keep questions and answers up to date. So if circumstances change and now a different answer is the correct one, accepting that/unaccepting the obsolete one is justified.

The answerer will lose 15 points from that, but the reasons for the unaccept are clear and understandable. Making the internet a better place by keeping the information up to date should easily outweigh those 15 points.

To avoid that the answer attracts downvotes, the easiest way would be to edit in a disclaimer that explains the situation. If it states that the answer previously was useful but now became obsolete, then nobody will feel misled.

share|improve this answer
    
Or better yet, to edit it into an answer that is now correct (if it's your own) or commenting to help the author bring it back up to date. –  Servy Jul 10 '13 at 14:27

Here's an answer of mine (How to Style a Checkbox Using CSS) that fell into this category -- the widespread availability of CSS3 functionality meant that my original answer became invalid. Someone pointed this out in a comment. I left the original text but edited to add a new answer. Interestingly, I believe I ended up with more upvotes after the new answer than I got originally.

So I concur with Aaron that I would comment on the accepted answer, both to warn other readers and to give the poster a chance to update their answer. If they don't, I'd eventually un-accept their answer and modify the question.

share|improve this answer
1  
Please note editing the answer with a version disclamer (instead of unaccepting/in addition to unaccepting) is also a good idea. –  Jan Dvorak Jul 10 '13 at 14:48

You must log in to answer this question.

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