How should the community handle the following problems related to current events?

  • Questions might become inaccurate/obsolete/wrong
  • Answers might become inaccurate/obsolete/wrong (even when heavily upvoted and accepted)
  • We need to monitor current events questions to see if they need updating

Please vote on the various solutions below (they should be independent) or add another idea for voting.

share
3  
Doesn't almost all information have the potential to become obsolete or wrong? Isn't it just a matter of degree? –  Andrew Grimm Apr 16 '11 at 7:45
2  
I agree, this is to handle particularly bad cases and exceptions based on quickly developing events. –  Sklivvz Apr 16 '11 at 10:18
    
Related question on meta.SO: meta.stackexchange.com/questions/88085/… –  Andrew Grimm Apr 20 '11 at 8:01

8 Answers 8

Make it a policy to add the following disclaimer banner to all current events questions:

> ![current event](http://i.stack.imgur.com/e8DCX.png) **This question is about a current event.** Information may change rapidly as the event progresses.

which looks like

current event This question is about a current event. Information may change rapidly as the event progresses.

OPs are expected to add it themselves, high reps and mods are expected to put it in if necessary.

share
2  
With this in addition to comments that already look like {{Citation needed}}, we'd really start emulating Wikipedia. Not that that's a bad thing. –  Jason Plank Apr 17 '11 at 4:13

Make it a policy for high rep users and mods to modify existing answers to add the following banner to answers that have become outdate/incomplete/obsolete.

> **The information in this answer is no longer accurate, events have progressed further after this answer was written.**

which looks like

The information in this answer is no longer accurate, events have progressed further after this answer was written.

share
5  
low rep users can either propose an edit or flag for moderator attention to achieve the same result. –  Fabian Apr 15 '11 at 10:33
2  
Just as you need to source an answer to prove it, don't we need sources to unprove an answer (assuming it was accurate/sourced/proven at one point in time)? Otherwise, users with questionable motives, an agenda or a vendetta could use this label as an attack. (Yes it can be undone easily, but those who would don't necessarily have eyes on every answer) –  NickC Apr 15 '11 at 19:04
    
@Renesis, that's why there is a reputation barrier of 1,000 for editing (and higher level users do get to see a list of recent edits). –  Sklivvz Apr 15 '11 at 23:42

If the highest voted or accepted answer is:

  • Mostly correct
  • In need of minor adjustment
  • And/or simply old, behind, or not completely developed given recent changes

...then propose edits to it, even substantial ones, to bring it up-to-date. This is, I think, what Joel was talking about and I've seen several answers (example) improved drastically on Skeptics by third-parties even when the topic wasn't rapidly changing. I am really encouraged to see users doing this.

If the highest voted or accepted answer has become totally wrong (in other words, events have come to light that have reversed what was believed or even what was accurate before), substantially wrong, misleading, or otherwise unsalvageable:

  1. Down-vote the answer.
  2. The "locking in of votes" could be a problem here (for users who already up-voted the inaccurate answer), so in addition, we should edit the answer with a notice about the inaccuracy of the answer. This notice should reference sources discussing the changes that have made the answer inaccurate. Any edit to an answer will allow you to vote again.
  3. If you can, post your own answer or vote up answers that are correct.
share

Stack Exchange sites are about getting expert answers, not for taking stabs at questions for which answerers do not have the sufficient knowledge to answer (i.e. because the event is rapidly changing).

Questions that ask about existing events should be considered on a case-by-case basis for the following:

  • Is the event sufficiently notable that answers will be interesting in the future? If no, the question should be closed as too localized.
  • Is the question even potentially answerable? If no, the question should be closed as not a real question.

In addition, answers should be considered on a case-by-case basis as to whether they actually answer the question. Guesses and speculation based on a lack of detail about the event are bad answers, and should be down-voted (or even removed). It's better to leave a question unanswered until it can be definitively answered than to provide a bad answers for the sake of answering it.

The alternative is for answerers, who don't have all the information but welcome others filling in the details, to mark their answer as community wiki. This significantly lowers the bar to contribute to providing a useful answer about an event, and is the closest analog to how Wikipedia handles current events. One person won't have a definitive answer about a current event, but potentially several people working together will.

share
    
One could say that answers about currently changing events are almost inevitably automatically bad for Skeptics since we require reputable references. Clearly incomplete and/or changing evidence is not good enough to support answers. –  Sklivvz Apr 15 '11 at 23:40

Apply a meta-tag to the question:

current-event

so that such questions can be easily monitored and checked.

share
    
That seems like a really bad idea –  Ivo Flipse Apr 15 '11 at 11:58
    
Agreed that it makes it easy, but I'm not a big fan of meta tags. It doesn't seem to suit their purpose. However, I am not 100% convinced that I'm right. –  NickC Apr 15 '11 at 19:06
    
I like this idea, could be used in addition to banners. –  user288 Apr 15 '11 at 19:54
1  
I think that if banners were a SE feature (like a flag on the question that gets rendered as a banner), then a search could be implemented without meta tags. –  Sklivvz Apr 15 '11 at 19:57

As an alternative to the options Sklivvz posted in answers, we can also Lock out of date questions to essentially archive them. Adding a banner to the top of the question body would be pretty clear. People can (if they see fit) re-ask the question now that the old question and its answers are out of date.

share

Delete questions and answers that have become obsolete. For example delete answers that have become wrong even if accepted and upvoted.

share
    
I can't think of an example of a question that can become obsolete. Answers, yes. Questions, no. –  Kit Sunde Apr 15 '11 at 20:32
5  
@Kit "Is it true the end of the world will take place on 12/21/2012?" Not that the question should be deleted, but after that date, the question is definitely obsolete. –  user1755 Apr 15 '11 at 20:42
    
@Mark Trapp - Yeah okay, that would count, haha. –  Kit Sunde Apr 15 '11 at 20:49
2  
@Mark - it may not need to be deleted - the world may have ended :-) –  Rory Alsop Apr 19 '11 at 9:58

Only normal SE guidelines are needed to keep the posts up to date and no other guidelines are needed for Skeptics

share

You must log in to answer this question.

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