What to do about old questions whose answers are now invalid

Question

This is inspired by the close vote on this question, which asks whether one should put MOOCs on one's CV. The comments are:

I would note that this was asked three years ago, here; I'm not sure if the answer hasn't changed a small (very small) amount since 2012 though, particularly as several MOOCs move closer to a kind of accreditation. Joe

[I]t is a duplicate. Etiquette is not to re-ask an old question for which the answer might have changed over time; but rather to add new answers, or update existing ones, on the old question. [...] EnergyNumbers

While I agree with the points raised, there are several problems here.

1. By construction, the asker of the new question cannot add a new answer to the old question or update an existing answer: they are asking because they do not know the answer. Is there anything that a user can do to encourage new answers to old questions that has approximately the same force as adding a new question?

2. Even if somebody does post a new answer to an old question, that new answer won't be noticed if there are high-scoring out-of-date answers. How can new answers to old questions get noticed? e.g.,

   +157 Giant lizards are the most important creatures on earth. – A. Dinosaur [200M years ago]

   1 Homo sapiens is having a big impact. – Hugh Man [13 mins ago]

3. Extensive editing of old answers seems misleading. People presumably upvoted A. Dinosaur's answer because they thought it was correct, not because they thought that anything he might change it to in the future would be correct. An unscrupulous dinosaur could, for example, change his answer to "The world is, in fact, controlled by a cabal of blueberries" which, now, apparently 157 people agree with. But, even after a reasonable but substantial change, the score no longer represents the community's view of the current answer. What should we do about high-scoring answers that are no longer valid?

Answer 1

I would suggest a perhaps rather unusual solution that hacks the SE model: for a case in which the situation has radically changed over time, create the new question and link to the old question as a possible duplicate with the explicit declaration that the new question has been created because the situation has changed. Then:

• If the community thinks things have really changed, close the old question as duplicate and add an edit at the top saying: "This is how things used to be, but see the new question because they have changed"

• If the community things things haven't changed, close the new question as duplicate.

Answer 2

Another possibility, similar to what jakebeal suggests, is to simply ask directly if the answer is now invalid.

This is more or less the approach I took with my recent question: How has the application review process for NSF graduate fellowships changed?.

It seems to have been well accepted by the community.

Of course, this could lead to questions whose answer is simply "No", but no one said all the question of our site had to be interesting.

Answer 3

Adding a bounty to an old question asking for new answers will likely get it noticed. New answers to old questions get added to a special review queue, which increases the visibility. As for answer that are now outdated, adding a comment saying they are outdated is probably useful or even an edit which explains how thing shave changed and why the answer is now out dated.