1
$\begingroup$

I have understood that if a question has community wiki (CW) status, then all answers must have it, too. I came across an exception in a question about refereeing. The answers by Chris Woodward and Joseph O'Rourke are non-CW, but all other answers to this CW question are CW. Is this supposed to be possible? Under what circumstances is this known or supposed to happen? It looks like a bug to me.

I have a guess why this happened: Joseph posted his answer at Aug 24 '10 at 23:25. He started editing his answer before the question was turned CW at 23:31, and the edit to the answer was saved at 23:44. The edited answer was saved as non-CW because the answer was non-CW when editing started, and this overrode the CW status inherited from the question. For Chris the time order is the same, so the same mechanism is possible. Can someone confirm or refute this guess? (Edit: This turns out to be false. It has nothing to do with edits. See the answer below.)

I feel I should add that I'm not upset by anyone receiving points for an answer that I think should be CW. I'm just curious.

$\endgroup$
3
  • 1
    $\begingroup$ I'm turning this into CW now. $\endgroup$
    – Todd Trimble
    Commented Apr 21, 2015 at 19:19
  • $\begingroup$ @ToddTrimble, thanks! In the future, should I flag any CW question that has non-CW answers to make the thread uniformly CW? $\endgroup$
    – Joonas Ilmavirta
    Commented Apr 21, 2015 at 19:25
  • 1
    $\begingroup$ Yes, please flag, although I really think these should be examined on a case-by-case basis. There may be (rare) exceptions to a general rule of uniformity. $\endgroup$
    – Todd Trimble
    Commented Apr 21, 2015 at 19:48

1 Answer 1

Reset to default
2
$\begingroup$

The reason is, as you assumed more or less, that the CW-status of an existing answer-post is not affected by the question being turned CW by OP of question. (The bit about editing is not relevant though; the key-point is whether the answer was given before or after the question was made CW, and how the CW-ing happened.)

I should rather say "was" than "is" as now this simply cannot occur anymore (as users cannot turn question-posts into CW anymore, and a moderator changing the status will turn the answer-posts into CW, too).

This was AFAIK "by design" as it would seem undesirable that some users actions on the question-post could have an effect on an existing answer-post, even more so as CW used to be irreversible. (Moderators had the ability to turn the full thread CW.)

If you come across such mixed-threads for CW-questions, it can make sense to flag for moderators with the request of turning everything CW. (I did so, with positve feedback, from time to time.)

$\endgroup$
4
  • $\begingroup$ I had no idea it made a difference whether it was the OP or a moderator that turned the question CW. I see now that the rest of the answers to the linked question were posted after the question turned CW. My guess was that editing answers at the wrong time caused this, but maybe this was not the case. $\endgroup$
    – Joonas Ilmavirta
    Commented Apr 21, 2015 at 19:02
  • $\begingroup$ The editing is not relevant (I will add this in the post). A user turning question into CW only had an effect on future answers, not existing ones, and this is what happened in this case. (Certain legacy forms of auto-CW-ing after too many edits might have had the same effect.) $\endgroup$
    – user9072
    Commented Apr 21, 2015 at 19:10
  • $\begingroup$ Thanks! Do you know if all these properties of CW posts are documented somewhere? The situation is a bit confusing, especially due to features varying in time. $\endgroup$
    – Joonas Ilmavirta
    Commented Apr 21, 2015 at 19:20
  • 1
    $\begingroup$ A good source of information is normally the relvant Meta Stack Exchange FAQ in this case meta.stackexchange.com/questions/11740/… There it is also possible, albeit not convenient, to trace the developpment over time checking the revisions of the post. See for example for some info supporting what I said meta.stackexchange.com/revisions/11741/4 [Note however that the timeline of changes for MO is different as for yearse MO ran on a "frozen" version of the software.] $\endgroup$
    – user9072
    Commented Apr 21, 2015 at 19:31

You must log in to answer this question.

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