10

As technology evolves, answers to some questions might get outdated. Now, if I see a question which I know has outdated answers, I can either post a new answer, or ask a question which points out why all the old answers are no longer valid and ask for newer answers.

But what if I don't know whether or not the answers to a 10 year old question are still valid and relevant? Am I allowed to post a new question asking if those answers are still correct? Or is there some other way of doing this?

Note: simply commenting on the answers will not work, since the OP is often no longer active on the site.

5

No. The whole point of the site is to provide authoritative answers. If the answers need to change, then new answers need to be added, not new questions added.

That way, even the historically-correct-but-currently-incorrect answers provide context.

Plus, the SEO on those questions can remain pointing searchers to the most up-to-date info instead of the old info.

Add new answers, place a bounty on the question to bring focus to it again, etc.

We tend to close questions asking if older questions are still relevant. You can bring them up here in meta, or even better, head to the DMZ and ask there.

10
  • 1
    +1 Should we be flagging these questions for mod attention, for example to remove the Accepted checkmark from a now-incorrect answer? Mar 3 at 18:56
  • Thanks! That makes sense. From this MSE post it seems other sites have a similar approach. One drawback is that new users will be unable to draw attention to questions they are interested in. I'll wait a bit before accepting this answer just to see if anybody else has something more to say.
    – nobody
    Mar 3 at 19:31
  • @nobody yes, that is the unfortunate effect of the policy. But we tend to close them and try to bring attention to the old questions when it makes sense.
    – schroeder Mod
    Mar 3 at 22:06
  • @MikeOunsworth I haven't found the "unanswer" function. I could have really used it in a couple of instances (especially when the OP dropped their account).
    – schroeder Mod
    Mar 3 at 22:07
  • @schroeder That seems like an omission from the mod tools if mods can't change accepted answers; at least on answers older than X days. Should that be a feature request to the company? Mar 4 at 13:55
  • @MikeOunsworth meh - given the ethos here that it's the contributor that needs to be happy, not the mods, I'm not sure that request would go through. Remember that wrong answers ought not to be deleted.
    – schroeder Mod
    Mar 4 at 15:48
  • I wonder if there's an argument to be made that in security, now-incorrect answers are actually dangerous, so we need the ability to clean them up more than other communities? Again, I'm just talking about being able to remove and re-assign the Accepted checkmark. Mar 4 at 15:53
  • 3
    @MikeOunsworth you're preaching to the choir on that one. I still don't have high hopes of the function being made available. The community should vote the right answer. A couple of times, I have placed a mod note at the top of the answer when it was clearly and unambiguously dangerously wrong.
    – schroeder Mod
    Mar 4 at 15:56
  • @MikeOunsworth accepted answer only means that "the asker was helped the most with that answer", not "this is the best/most up-to-date answer".
    – Andrew T.
    Mar 11 at 11:33
  • 2
    @AndrewT. Yeah, I get that. That's generic text that's on the help page of all SE sites. With programming or hardware sites, an answer from 4 years ago may not work with latest libraries or hardware and you'll find out pretty quickly if it doesn't compile or run. The problem with security is that an answer from 4 years ago may "work" just fine, but be totally insecure because some groundbreaking attack was discovered in the last 4 years. I'm arguing that makes security.SE (and crypto.SE) different from other SE sites. Mar 11 at 15:25

You must log in to answer this question.

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