tea.mathoverflow.net - Discussion Feed (unanswered questions)

Anton Geraschenko comments on "unanswered questions" (5474)

Anton Geraschenko — Sat, 08 May 2010 18:16:50 -0700

By the way, another question which is closely related (and still unanswered) is Dave Brown's question Can an algebraic space fail to have a unviersal map to a scheme?. I'm pretty sure this is equivalent to the question, "can it happen that an algebraic space colimit exists, but a scheme colimit does not?"

Martin B. comments on "unanswered questions" (5473)

Martin B. — Sat, 08 May 2010 16:16:59 -0700

The colimit problem is solved now (by Anton and Brian) :-). Anyway, there are other unsolved questions and I think the discussion so far gives good ideas how to handle this ;-).

@Anton: I hope I can "reward" you by solving your old question. Today I spent some hours on it.

Anton Geraschenko comments on "unanswered questions" (5466)

Anton Geraschenko — Fri, 07 May 2010 21:14:40 -0700

I'm really surprised that question doesn't already have a satisfying answer. I'm going to email somebody who almost surely can answer the question. Incidentally, one of my old questions is related and also doesn't have an answer yet.

In general, I think breaking down the question into smaller questions and posting those is the right strategy, but in this case, I think the question is at exactly the right level of generality. So I think the right thing to do is to publicize it here. I'm opposed to posting a duplicate.

I agree that the bounty system is broken. I've been pushing for the solution proposed here, but I think it will probably take a while before it can be implemented.

Harry Gindi comments on "unanswered questions" (5464)

Harry Gindi — Fri, 07 May 2010 15:23:47 -0700

I also suggest that you give a link to this meta discussion so people don't try to close it as a duplicate.

Scott Morrison comments on "unanswered questions" (5462)

Scott Morrison — Fri, 07 May 2010 15:19:30 -0700

@Martin B.,

I'd second the proposal to ask a related, hopefully more detailed, question. You can see a fabulous (although unintentional!) example of this in my question, which didn't get very many answers, and then Noah's question on a very similar subject. The second question got an answer that we turned into a paper!

Martin B. comments on "unanswered questions" (5455)

Martin B. — Fri, 07 May 2010 12:14:20 -0700

as I already mentioned in the comments, I've made a bad experience with a bounty. an answer was accepted due to the time limit, although it was no answer but rather a comment how to interpret the question ....
The second time I used a bounty was a very easy problem and it worked well.

Harry Gindi comments on "unanswered questions" (5454)

Harry Gindi — Fri, 07 May 2010 12:05:48 -0700

@Everyone who is suggesting anything other than re-asking: If one edits a question or offers a bounty on a question that has an answer with nine votes, there will be no new answers (I speak from experience). I feel like the only reasonable way for this question to get an answer is for Martin to ask it again, perhaps with clarification why the answers in the other thread were not sufficient.

I think that this is the best way to go about it, as long as the administration has no objections.

Ben Webster comments on "unanswered questions" (5453)

Ben Webster — Fri, 07 May 2010 12:05:47 -0700

Martin- This is exactly why bounties exist.

Of course, you could also rewrite the question. As a general rule, if people keep give you answers that aren't what you want, that's a sign that the question isn't clear.

Steve Huntsman comments on "unanswered questions" (5452)

Steve Huntsman — Fri, 07 May 2010 11:57:50 -0700

Offer a bounty.

Martin B. comments on "unanswered questions" (5451)

Martin B. — Fri, 07 May 2010 11:53:48 -0700

thanks. that's a good idea.

Harald Hanche-Olsen comments on "unanswered questions" (5449)

Harald Hanche-Olsen — Fri, 07 May 2010 10:56:00 -0700

It's way outside my field of expertise, so I can't comment on that aspect. But your question has gotten quite a few votes, plus it has attracted a lot of attention. If you did not get a satisfactory answer then I'd venture a guess that answering it as asked is actually quite hard, and nobody here knows the answer. In which case simply editing it in order to get it bumped to the front page won't buy you much. Maybe you could ask a new question that is related and would somehow shed light on the original question? Or you could rephrase the original a bit trying to make it clearer what exactly you want to know, especially in light of the answers that have already come in.

In summary, I'd say that editiding a question for the sole purpose of bumping it to the front page is bad form. But asking the same question anew is worse, and will likely get the new question closed as a duplicate. But if you put some work into it as outlined above, it's a different matter altoghether. Then you are actually conducting a conversation about your problem, and that is good.

Martin B. comments on "unanswered questions" (5448)

Martin B. — Fri, 07 May 2010 10:50:18 -0700

Well it only bumps it for the people who look at active (instead of new) questions. Are they the majority?

To be honest, the answers so far did not help.

Qiaochu Yuan comments on "unanswered questions" (5446)

Qiaochu Yuan — Fri, 07 May 2010 10:42:33 -0700

Editing a question bumps it. You don't have to add any new content, although you might want to summarize what you've learned since posting the original question.

Martin B. comments on "unanswered questions" (5443)

Martin B. — Fri, 07 May 2010 10:25:52 -0700

I've just signed in here for this question.

There are some questions which I've posed on MO which received no answers or at least some answers which contain interesting facts, but don't actually answer the question. One Example:

http://mathoverflow.net/questions/9961/colimits-of-schemes

Now how can I make this question come to the fore? My mathematical problems are not "serious" in the sense that I don't need them for an official research project. Nevertheless, the question above is important for me and I have not been able to solve it by myself. And there is a bunch of questions in the comments ...

I've already made the question active artificially (answer something and delete it right away), this produced at least a comment of Brian.

Shall I start a new question? Is this acceptable?