tea.mathoverflow.net - Discussion Feed (Closing answered but lingering questions)

Andrew Stacey comments on "Closing answered but lingering questions" (6193)

2010-06-18T01:36:51-07:00

It is important to remember that the periodic stirring that the MathOverflow User does is intended as a good thing: to bring back to our attention questions that, as far as it can tell, haven't been resolved. As with any automatic process, it sometimes gets it wrong, so now we know how to help it: ensuring that there is an answer with a positive vote count. If there is no suitable answer, but somehow the question has been resolved (maybe in the comments), then a community wiki answer explaining the situation is more than likely to fix it.

CSiegel comments on "Closing answered but lingering questions" (6192)

2010-06-18T00:00:46-07:00

I'm fairly certain that this happened to one of my questions, which is why it now has a partial answer. Thank you Mathoverflow user!

VP comments on "Closing answered but lingering questions" (6184)

2010-06-17T20:44:39-07:00

Scott and Joseph: I came to this thread to share my wonderful solution to the mystery, but I see that I wasn't the first :( In fact, I think this happens rather often. I, too, noticed more than a few questions that have been raked up by "Mathoverflow" and had positive votes when I looked at them. However, they are usually in the middle of the stack, so presumably, they have been sitting on the front page for several hours by then.

Joseph O'Rourke comments on "Closing answered but lingering questions" (6178)

2010-06-17T11:21:54-07:00

Scott, I now see that is a possibility. I had initially rejected that as a possibility because (1) The answer was 2 Jun, and (2) The MO bot modified it yesterday, two weeks later. But perhaps there were two votes for the answer after the bot mod and before I looked at it. It is logically possible. I hadn't thought of that. Actually, the bot mod might have brought it to the attention of the community and caused those votes. So yours is a plausible explanation.

Scott Carnahan comments on "Closing answered but lingering questions" (6174)

2010-06-17T09:58:02-07:00

Joseph, it seems possible to me that someone voted the answer up between the time the community user bumped the question and the time you looked at it. Are you sure the answer had been voted up before the question was bumped?

Joseph O'Rourke comments on "Closing answered but lingering questions" (6150)

2010-06-16T13:17:25-07:00

Thanks for checking, Anton. I believe you. Best to let this little mystery lie. There are much larger mysteries out there awaiting our attention. :-)

Anton Geraschenko comments on "Closing answered but lingering questions" (6148)

2010-06-16T12:56:18-07:00

The community user's actions aren't logged and are pretty hard to track. As far as I know, the community user "pokes" questions that haven't had any activity in a while and are "unanswered". The SE lead developer confirmed what "unanswered" means. See also this meta.SE post. I'm not sure what happened here.

Joseph O'Rourke comments on "Closing answered but lingering questions" (6133)

2010-06-16T06:30:26-07:00

Yes, it appears I strayed into the heavens where mere mortals should not cavort!

Andrew Stacey comments on "Closing answered but lingering questions" (6132)

2010-06-16T06:25:17-07:00

Hmm, according to Anton's answer on meta.stackexchange, that shouldn't have happened as David Roberts' answer got voted.

At this point, I'd wait for Anton or Scott to wake up to give you the definitive answer on how the community user is meant to behave and whether or not there's something else going on behind the scenes that us mere mortals can't see.

Joseph O'Rourke comments on "Closing answered but lingering questions" (6131)

2010-06-16T06:03:03-07:00

Andrew, you may not be right re bumping? I just noticed
http://mathoverflow.net/questions/26699/double-category-of-topological-stacks
which has been bumped by the MO process but has a voted answer...

Joseph O'Rourke comments on "Closing answered but lingering questions" (6130)

2010-06-16T05:53:14-07:00

Thanks, Andrew, I did not realize two MO features: That the bumping only occurs with no-voted answers, and that one can designate an answer community wiki. So I followed your advice with the 3D polygon problem, and we'll see if that resolves at least that one. If so, it gives a path for resolving other similar situations. Although the resolving mechanism is sufficiently obscure that I doubt it will catch on.

Andrew Stacey comments on "Closing answered but lingering questions" (6129)

2010-06-16T05:26:42-07:00

Someone who knows better should confirm this, but I think that questions with an answer that has been voted for don't get this bumping treatment (the two on the question in question - I love writing that! - both have no votes) so a fairly innocuous solution is simply to vote for one of the answers.

If none of the answers deserves a vote, or if there are no answers, then you could provide one. In particular, if it has been answered in the comments then that can be transferred to an answer. Custom dictates that such an answer should be community wiki. You would then have to hope that someone else voted for the answer (you could even write that in the answer: "I'm writing this answer to stop it getting bumped to the front page, I need someone to vote for it for my nefarious plot to succeed"). This is what you say in your last paragraph with the addition of the "community wiki" feature.

Of course, this presupposes that the question shouldn't have been closed for a genuine reason (wrong level for the site or some such) in which case vote to close (and/or bring it to the attention of the +3000 users by commenting on it here on meta).

Indeed, some people only answer in comments (something that I consider a bit unhelpful to the community) and I see no problem with copying the answer into a (suitably attributed and community wiki) answer for the questioner to accept.

Joseph O'Rourke comments on "Closing answered but lingering questions" (6128)

2010-06-16T05:04:41-07:00

http://mathoverflow.net/questions/17048/compute-the-centroid-of-a-3d-planar-polygon

As a relative newcomer, I am noticing something of which you long-timers must all be aware:
Some questions are pretty much closed in the comments, or even in the answers, but they linger on,
either because the poser can't select comments as answers, or because the poser is not attending.
And the random MO background process keeps bumping them to the top of the active list.
I notice it with this 3D polygon question because that is a perfect topic for me, so it catches my
eye, but it is effectively closed and has been for two months. Just yesterday JBL complained about it lingering on.
I'm not offering a solution, just pointing out a problem.

Concerning the "closed in comments" situation, I wonder if it is possible for the poser to answer their own question (by reformulating the comments) and then accept their own answer? This might solve some of the lingering problems, but
those lingering due to inattention or other reasons would continue to linger. Letting them languish forever would be fine
except for the process-bumping.