5

The accepted answer for this question points to a Chrome extension. The answer was correct in 2009 but Chrome has moved on and now this answer is more correct.

I have upvoted the seemingly more correct answer and I am considering down voting the now-outdated accepted answer but it seems a shame to penalize the posters for answers that were once correct but are now dated.

Is there a better way?

7

The answer isn't worse because it only applies to older version of Chrome. The question doesn't specify a Chrome version, so answers for any versions are equally valid. It's not either user's fault that the versioning scheme of Chrome makes it impossible to tie questions to versions, like we'd do with e.g. .

I don't think downvoting is the way to go here. We wouldn't downvote a late answer that only applies to older software versions either (something that should be actively encouraged! It's often the older versions of software that require the tricky workarounds). The only difference would be that the later poster can already know for which more recent software versions his solution will not work anymore.

The answer has a built-in warning sign in the date and time it was posted (Chrome + 3 years old information = irrelevant), and it's possible to sort answers by date, allowing readers to start at the most recent ones (admittedly at the bottom).

Instead you should edit the answer to provide more information about the assumptions of the answer ("In Chrome N.0 and earlier, ...") or add a comment stating "As of version N, this answer no longer applies". This way, readers will be able to quickly determine which of the answers applies to their situation.

The more recent answer should rise to the top over time anyway, as future readers will find it more relevant.

  • 3
    editing seems reasonable. As for sorting -- the average seeker who finds SU may not look past that big green checkmark. – Jeremy W May 31 '12 at 18:06
  • 3
    I'm just imagining how I browse Stack sites when I'm looking for help. I'm always annoyed by accepted answers that stick at the top but are terribly outdated. The whole point of SE is to keep signal to noise ratio high and in that case The "built-in warning sign" doesn't help here if the answer just happens to become incorrect. – slhck May 31 '12 at 21:57
  • 1
    @slhck "just update your production system to the nightly build" cannot be the answer either. People are still running Firefox 2, windows xp, and rhel 4. Voting "this answer is not useful" is misleading. Edit, or add a comment, and move on. – Daniel Beck Jun 1 '12 at 5:53
  • No one mentioned nightly builds. But the question is who you want to target. Of course editing makes sense, but people don't do that too often. – slhck Jun 1 '12 at 7:07
  • @slhck If you have the time to downvote, you have the time to write "This does not seem to work in $currentVersion anymore." / And many organizations are very slow in updating, even to regular release versions. "Nightly build" might be exaggerated, but if you need to use an older version, you're thankful for answers applying to them. – Daniel Beck Jun 1 '12 at 17:56
  • Yeah, of course I'd mention that something doesn't work or post a better answer instead. It's just that it's hard to get the newest / most correct information to float to the top faster, especially for old and popular questions. But that can probably only be solved by using a metric other than sorting by the accumulated net score :/ – slhck Jun 1 '12 at 18:45
  • It looks like some of the answers to the old question have already been cleaned up. I think constructively editing the accepted answer is a great idea. – Jeremy W Jun 4 '12 at 13:31
  • Wikipedia addresses a similar problem with an "as of" template. If information is of the sort likely to become outdated, editors are encouraged to use this template - later editors can easily find information that has been tagged this way (or which has an "expiration date" which is passed). On the one hand, in technology most information eventually expires. On the other hand, we have version numbers on everything so we can often handle expiration. But maybe an "as of" feature would be helpful on SE for cases where information is likely to expire in an unclear way? – Shannon Wagner Jun 11 '12 at 19:06
  • @ShannonWagner Some kind of "vote to mark as superseded/outdated" that adds a notice to a post might be a good idea. I proposed that in a similar topic a long time ago but it wasn't very visible. It might be time to discuss for a wider audience, but I'm not sure if that problem is widespread on other sites. Non-tech topics aren't usually changing that fast... – Daniel Beck Jun 11 '12 at 19:09
3

You should leave a comment stating the answer is old, and needs to be updated. If the author hasn't updated the answer, there's nothing wrong in downvoting the answer - the penalty associated with it just might force the author to update in case the author's not updated the answer despite leaving the comment.

  • Depends on whether the authors are still active on the site, which often isn't the case for really old answers that just happened to stick around. I would even vote for outright deleting those that just convey wrong information if there's a (now) correct answer around, if the author had the chance to improve their post or delete it at their own will. – slhck May 31 '12 at 21:55

You must log in to answer this question.

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