I don't really see the problem with individuals being (more) elitist (than others). It is not particularly rude, and I doubt many of those questions come from people who have given us the courtesy of reading the FAQ. That said...

I usually don't write "MO-level". I think the wording I prefer is that "... your question does not fit into the scope of MathOverflow." Or sometimes "this problem is ill-suited for MathOverflow".

MO is elitist. That's a feature, not a bug.

I have been contemplating this issue for a little while, and I am coming around to thinking that the average person who posts a non-research level question to MO does not know what we mean when we say "this site is for research-level questions", i.e., they do not know the concept of a research-level math question.

Note that I said the concept, not just the terminology: if you asked them to describe to you what kind (or level, or whatever) of problems a research mathematician works on, they would either have no idea or very mistaken ideas.

So I don't see an easy fix here: we are trying to tell people that this site is for math questions at a level beyond that which they have any prior experience or real conception that math questions can be. (Now there's an elitist sentiment! But it seems to be true.)

Guys, as a slightly tangential comment, can I make a gentle and selfish request of the people who have voting power?

If you are the person to cast the final closing vote on a question for the reason that it is not MO-level (or in general for being of no interest to research mathematicians at all), would you also be so kind to re-tag the question to [tag-removed] so those of us who filter that tag can benefit from your efforts?

I know Yemon and I (and possibly some others) have been removing tags for such obviously inappropriate questions when we see them on the front page, but this will of course bump them a little bit and cause them to stay longer on the front page. It would help if the re-tagging happens closer to when the question is closed. Thanks!

@Mark: Thanks for pointing that out. I realize now that I had read Nate Eldredge's post, and it has been rattling around in my head ever since. As we mathematicians sometimes say, I acknowledge his priority. :)

Steve, I would probably stop doing things that required me to take online calculus tests to participate...

I like Steve's suggestion. And I think Mariano is beyond the rep threshold so that he likely wouldn't have to worry about solving too many of these problems.

I agree with Mariano. I think we should be as welcoming as possible to mathematicians who haven't used this site before, and a good way to do this is to remove as many hurdles as we reasonably can. I don't think the rate of low-level questions is substantially worse than it was 1 year ago, but if we reach some kind of crisis state, you should feel free to contact the moderators.

@Ryan, to use a recent example: Tate was not beyond the rep level (!!)

@Mariano, maybe I don't have the best context but I'm guessing that wouldn't be a real barrier to entry to pretty much any research mathematician. True some of us maybe haven't taught or thought about calculus for years.

Perhaps an alternative would be to raise the bar for what it takes to ask questions. Giving answers would still be open to all. So that wouldn't stop people like Tate. But to ask a question we could have a naggy pop-up perhaps with questions that you get to avoid only when you're past a certain rep. So Tate would presumably never experience this, since he'd get sufficient rep on his 1st answer.

Gerhard, in some sense having any standards for posting is elitist, isn't it? To post a research-level mathematics question requires some level of knowledge of what that means, which for many people is rather inaccessible information and of course there's a wide gray area -- there's something of an active attempt to make that gray area a little less gray, by sending certain types of math questions to other forums. I believe the Tate example was meant as an "extreme point among possible negative consequence of introducing barriers to posting". As far as I can tell, adding Tate's name was just meant to emphasise the point, via a recent example.

Dear Gerhard,

Tate won the Abel prize earlier this year. I think that people can be justifiably proud that an Abel prize winner has participated on MO (and that Fields medalists also participate); it is (not the only, but one) measure of success for the site, and it is reasonable to try to think about what aspects of the site contributed to this success, and (at least) be sure not to damage those aspects too much.

Regards,

Matthew

I actually don't think the phrase "MO-level" is impolite. It may be elitist, but I dislike using it for a reason similar to that Pete Clark brought up (and which Nate talked about in another thread) that it conveys absolutely no information whatsoever. Whereas the phrase "research-level" does not necessary have a definite meaning for amateurs, it at least establish a (possibly quite arbitrary) criterion for inclusion. With "MO-level", the logic is circular: "your question should not appear on MO because it should not appear on MO."

Whether or not we ever reach a conclusion on this elitist/politeness business, I think the phrase "MO-level" should be eliminated for at least the argument above.

Now, about politeness: it is a lot easier to offend on the internet. And it is also a lot easier to take offence on the internet. This is of course due to the oft-observed triviality that the textual nature of the communication eliminates contextual clues, which we sometimes try to reinstate by inclusion of smileys. Since this cuts both ways, I think, in general, barring overt rudeness, it is hard to establish a "suitable level of politeness", especially considering that individuals of different generations tend to have different perceptions on what is and is not appropriate to say on the internet.

Which is all just my way of saying that I disagree with the premise that MathOverflow should have a "polite interface" with the public, where "polite" is taken to mean "smooth" or "showing regard" or "deferential to". I agree, however, that we can be at least "polite" in the sense of being "civil". Which means that while I find it unacceptable to say a question is too "simple" or "easy" or "trivial" for MO, I also don't see it necessary to sugarcoat our message that MO is for discussing mathematics one is likely to encounter in one's post-graduate mathematical education and thereafter.

(An aside to finish this longish post: @Gerhard Paseman: you do mean the "Mr. or Ms." in a tongue-in-cheek way, right?)

Yemon: oh. I misunderstood your meaning when you used the word "offset". I thought you meant "create an offset" or "cancel". (As in, say, the money people donate to offset their carbon emissions.)

Evidently we agree on the actual point. I'll remove that tangential edit above for clarity.

Yemon, I forwarded his two letters to you, the second includes my answer to his first.

For questions blatantly not "MO-level" (and here I use that blasted phrase again), or those whose existence does not contribute to the mission of MO, I would perfer if all other tags are removed so as not to pollute the tag-space. Then people actually refining their searches using tags will not have to deal with questions that clearly won't address their problem.

Thanks for doing that.

Gerhard: the possibly mistaken misreading that you don't apologize for, do you mean your possible misreading of previous posts, or my possible misreading of your post (on the John Tate point)? The meaning of the clause after the colon depends on what you mean there.

(I'm not trying to argue or anything. I am just trying to figure out what you want to say. I think you meant the first of the two options, but the phrasing is a bit ambiguous.)

Gerhard, i have Willie's email address (which i think may be obvious on his profile, I don't remember), and Willie, I have Gerhard's. I think an infinite regression can be avoided if you write to each other individually. In this case, it appears you have no actual disagreements.

If I remember correctly, there were no objections (the last time we had this discussion) to only allowing unregistered users to post answers. This seems like it would pretty much get rid of most of the nonsense/homework/garbage.

The problem was that it was (and still is) not currently possible to do this through the SE software (and therefore it would have to be hacked).

The text could be added to the FAQ, so that you need only write "Please see http://mo.org/faq#relevant-section".

I hope that you realize that Vaughan has not returned since he asked that question...

Also, even after reading the article, I don't see how MO has jumped the shark.

If we can get Grothendieck to start visiting, then MO will have jumped the shark!

In the form for asking a question, a list of possibly related questions appears after enough words are typed. Perhaps this can be used to raise an additional flag if the question appears to be related to too many closed questions? It could raise a window saying that "the question is likely to be closed because it duplicates or otherwise resembles other questions that were closed; please see the FAQ for what is likely to be closed" . If the right language is used, it may discourage many who trigger the window, without presenting an insurmountable barrier to those who have a valid question to ask (based on the perhaps faulty premise that one posing a research level question is likely to read the flag window, one or two related questions, and the FAQ before submitting, as well as on the additional perhaps faulty premise that there exists appropriately discouraging language to those who don't know what a research level question is) .

Gerhard "For Sale: Cheap, Quick Solutions" Paseman, 2010.10.14

Sorry -- I think that Ben used "jumped the shark" in a slightly unconventional way, as in "reaching a certain level of ridiculousness", without necessarily "going bad", and I reused the phrase as a memorable pointer to the particular incident, compounding the confusion.

The point remains -- if someone bothered to analyze whether various prominent mathematicians on mathoverflow registered before or after asking their first question, I'm convinced it would be clearer that such barriers are dangerous.

Actually, my supposition from seeing a few instances on MO was that, over time, the sense the phrase conveyed had simply changed. Since I had never seen the phrase before MO it was all fine by me, I did think people might want to know. I did not wish to cause discomfort. Compare:
http://mathoverflow.net/questions/42139/estimating-direction-from-a-distribution-on-a-circle/42342#42342

Will, I just stole your nice wording (and modified it slightly). I hope you don't mind!