Comments for Stack Exchange Mathematics Blog

Comment on Some Statistics on the Growth of Math.SE by Ian

Ian — Fri, 18 Sep 2015 11:57:28 +0000

Remember that down votes questions without answers get auto deleted based on their age. Therefore it will always look like newer questions are worse then older questions.

Comment on When can we do induction? by Remember me

Remember me — Wed, 05 Aug 2015 12:34:16 +0000

Great Blog!!

Comment on Some Statistics on the Growth of Math.SE by quid

quid — Wed, 13 May 2015 23:03:42 +0000

(It seems one cannot reply to a reply; this is intend as reply B.M.S.)

Whether you like it or not some things are rules, not only my personal preferences. These include: one must not flat-outright insult or denigrate other users (in any case not as standard form of communication), one must not vote in spite, one must not cheat with votes in other ways, etc.

I would already be happy just this would be follow. It should not be that hard; still several highly active users fail at some of it. (Granted "vocal" was perhaps not the most fitting adjective.)

Comment on Some Statistics on the Growth of Math.SE by Brian M. Scott

Brian M. Scott — Wed, 13 May 2015 09:55:06 +0000

The most vocal users, especially on Meta, are those with a rather narrow and often mathematically elitist view of what is appropriate, many of whom confuse their preferences with the rules of the site. Their efforts to moderate the site frequently end up imposing their prejudices on other users and not infrequently verge on bullying newcomers, especially those with elementary questions. And the campaign begun about a year ago by The Man of Many Names has resulted in out and out vandalism. We could do with a good deal less of this ‘moderation’.

Comment on Some Statistics on the Growth of Math.SE by quid

quid — Tue, 12 May 2015 23:43:13 +0000

Thanks for the detailed analysis. At a glance I do find the increase in unanswered question a bit worrisome. I only hope the efforts to moderate the site will continue and they will not be stifled too much by a group of some vocal users with an at times cavalier attitude towards the rules of the site.

Comment on Some Statistics on the Growth of Math.SE by Bye_World

Bye_World — Mon, 11 May 2015 19:23:49 +0000

@anorton Maybe you don't understand what I mean by "boring". But it would be entirely infeasible to implement a boring-questions-must-be-deleted policy. They make up way too much of the volume of the site AND there's too much room for opinion about what an "interesting" question is. Example: I "specialize" (I'm actually a physics major, not a mathematician) in Hilbert spaces and operator theory (really QM, but that's the math I mostly use). Should I downvote any question about matrix arithmetic -- even ones where the questioner has put in some effort and I think the concept is one that many students struggle with -- just because to me that subject is "old-hat"?

Comment on Some Statistics on the Growth of Math.SE by mixedmath

mixedmath — Mon, 11 May 2015 06:27:23 +0000

I do not mean to imply that I only vote on questions that I answer. I vote frequently. What I was trying to get at is that if I don't think a question deserves an upvote, then I won't answer it either. So not if and only if -- just a one-sided implication.

I agree that question quality and votes are poorly correlated. I might be biased, because I'm typically more interested in less approachable problems. Selection bias leads to their votes being less meaningful, I think.

Within a question, I think that votes do a pretty good job correlating with quality. But I think that great answers to less approachable problems are also doomed to fewer votes.

Comment on Some Statistics on the Growth of Math.SE by Brian M. Scott

Brian M. Scott — Mon, 11 May 2015 05:31:25 +0000

I’m inclined to agree. Besides, what counts as boring varies enormously from person to person: there are whole areas of mathematics that I find boring (e.g., complex analysis)!

Comment on Some Statistics on the Growth of Math.SE by Brian M. Scott

Brian M. Scott — Mon, 11 May 2015 05:29:20 +0000

Whether I upvote a question has almost nothing to do with whether I answer it. I am most likely to upvote a question as a reward for providing some constructive thoughts, even if they’re wrong. A very distant second reason is to provide a compensatory upvote when I think that a user has been unreasonably slammed. I have no idea what criteria others use, but in general I see very little correlation between votes on questions and anything that I can recognize as quality. I think that little if anything would be lost if questions were not subject to voting at all. (Note, however, that under the present system I do make a genuine effort to vote on them: 42% of my votes are on questions.)

In contrast, it appears to me that there is something vaguely (albeit sometimes very vaguely) resembling a consensus when it comes to voting on answers, apart from the vandals who downvote correct, usable answers on ideological grounds. My own judgements are often rather different, but the voting that I see on answers does seem to have a positive correlation with quality. (This can’t be pushed too far, of course: different types of questions have audiences of different types and sizes.)

Comment on Some Statistics on the Growth of Math.SE by anorton

anorton — Mon, 11 May 2015 04:26:21 +0000

When I'm interested in answering questions, I tend to hang around The Crusade of Answers. I wholeheartedly believe that our site is reaching the point where we will need to take drastic steps to optimize for pearls, or we will cease to be a high quality site.

That aside, I really enjoyed reading this post and thought it was quite informative. In particular, I look forward to hearing some of the results of the project dealing with computational linguistics.