# Why do non-research and soft questions get closed if and only if the poster lacks high reputation?

The title says it all.

Don't get me wrong. I have actually enjoyed my first few weeks on MO. I am not writing out of personal vendetta. That is, I have not had any of my own questions closed and I usually get positive feedback. I have probably not asked any closable questions (well, until now, where I am in jeopardy of having this one closed).

From a relative newcomer's vantage, it's clear that discrimination is applied when deciding whether to downvote or close such questions. I have seen similar-style questions survive (and thrive) when asked by a hot-shot yet closed when asked by a newbie.

For example, consider the "journals" category.

This question seems quite articulate to me, but it is the lowest score in the category and was closed (rightly) because it is too broad. Its topic is not specific to mathematics.

This question is also not math-specific, but generated a lot of interest and 155 upvotes. It was eventually closed as no longer relevant after a good ride.

It is easy to find more examples. Pick a softish category or search term and sort by votes.

I fully admit that those on MO with high reputation have earned it by being good (sometimes really good) at asking and answering. And many of the newcomers can't even form a sentence. There are examples (like the one above, though) where I can't explain the discrepancy in attitude. The reputation breaks as expected.

Does anyone take the view that MO is neutral in its treatment of off-topic questions?

edit: I think I am backpedalling a bit from my initial view (which, by the way, was never intended to be aggressive... more tongue-in-cheek, in fact). I can honestly say I see a correlation (a good word, Gil) between reputation-backed questions and forgiveness, but I now admit it's not a strong one. Just today, I noticed a high rep user getting trashed on metaMO because his question (which was fair) had a resounding and obvious answer (with which I agree).

(This is off-topic, but I suspect most of the downvotes he got were from people saying "NO!" to his question, rather than "YUCK!" to the quality of the question. I wish people would learn the difference.)

Anyway, maybe the issue is simply that a question cannot be separated from its asker. If most people on MO need to look at who's asking in order to make a judgement, then that's their right. I admit I do it. Our time is precious.

• If the intent is to vent/object, consider it noted. If the intent is to raise an issue for consideration, I suggest changing to a more neutral tone, especially the last paragraph. Otherwise readers are likely to " get you wrong ". – The Masked Avenger Mar 27 '14 at 19:45
• Just a few points. (1) I think a clearer discussion might be had if you provide a few examples. (2) With regard to the question of whether "MO is neutral", the way I see it, there is no definite entity called MO to which one could apply the word "neutral"; it's spread across many, many users, who make their own individual decisions. (3) If there is ever a question as to whether a question is treated unfairly, then meta is always a good place to raise it. Do you know about this thread? meta.mathoverflow.net/questions/223/requests-for-reopen-votes – Todd Trimble Mar 27 '14 at 19:45
• @ToddTrimble: Isn't it possible to argue that an aggregate entity is not neutral? Sociologists and politicians do this all the time. Thank you for the link. As mentioned, though, I don't have a particular beef with any closed questions. It's the off-topic ones asked by hot-shots that I wonder why they survive. I interpret the role of MO rather strictly. – Peter Dukes Mar 27 '14 at 19:59
• @MaskedAvenger: If I am honest, I think the goal is to vent/object. Todd's comment is accurate: I am a fool to ask that MO "change" somehow. Individuals will continue to take their own actions. You're right about the neutral tone, though. I did my best. – Peter Dukes Mar 27 '14 at 20:02
• I think "neutrality" is a pretty problematical concept here, despite what sociologists and politicians do. I'm pretty sure I don't care to defend something I don't put much stock in! But maybe I can get the ball rolling in an "answer" that gives one example that you might adduce for your position; there was some debate about it at the time. More examples can be added. – Todd Trimble Mar 27 '14 at 20:07
• I'm looking at your second example. I don't think it really supports the claim: the OP was very new to the site (and hadn't acquired a lot of MO rep) when the question was raised, and the question still got a warm welcome. (The closure reason "no longer relevant" is no longer one of the recognized stock reasons for closure, but was applied on MO.1 when a question had outlived its usefulness, e.g., when it had begun attracting crap answers.) – Todd Trimble Mar 27 '14 at 20:50
• Fair enough. It is not an example of reputation helping a question survive. I still sense inconsistency between the questions' treatment. Note, too, that high-reputation answers can "protect" an otherwise vulnerable question. – Peter Dukes Mar 27 '14 at 20:55
• The assertion that refereeing is not specific to mathematics is in my opinion false. The way refereeing is done in mathematics is very different to the way it is done in other scientific disciplines (perhaps not all other disciplines, but I think math is rather an exception than a rule here). – user9072 Mar 27 '14 at 21:31
• I agree. On the other hand, mathematicians have been at the forefront of pushing back against publishers, and this gives some merit (mot much, I admit) to the first question. Our relationship with publishers is uniquely strained, in my view. – Peter Dukes Mar 27 '14 at 21:46
• Dear Peter, did you check your assertion by looking systematically at the say newest 200 soft questions? Of course, for technical and for soft questions we can expect positive correlation between the quality of the question (as perceived by the MO community) and the reputation of the poster, but it is quite possible that this effect is weaker for soft questions. – Gil Kalai Mar 28 '14 at 11:07
• The question with 155 was asked in the old days when soft questions were more tolerated. – Benjamin Steinberg Mar 28 '14 at 12:55
• @PeterDukes, it is fairly common for new users to try to prescribe how MO works. I did that, years ago. Eventually, you find that MO goes as MO goes, it is neither perfect nor horribly corrupt, and you can get help and harmless entertainment here. Hmmm. If not entirely harmless, at least legal in most countries. – Will Jagy Mar 29 '14 at 5:53
• Peter, the history of interactions with that particular high-rep user has been fraught and contentious, and I suspect that some of the downvotes were not in the spirit of "no!" so much as "oh please; here we go again...". In other words, a particular form of "yuck", anticipating some emotional arguments to come. – Todd Trimble Apr 5 '14 at 21:15
• @ChristianRemling: True. I would like to believe, though, that identical words out of different mouths would get treated equally (should such an experiment be possible). – Peter Dukes Apr 7 '14 at 3:51
• *** Human nature *** – Włodzimierz Holsztyński Apr 7 '14 at 21:26

One could definitely argue that Peter Dukes has a point here. I recall one famous case, involving a post by Bill Thurston, who posted a "soft question" Thinking and Explaining. The question got a very interesting reaction, in terms of thoughtful answers surely, but also in terms of upvotes and the highly upvoted comment of Felipe Voloch: "What I am really thinking is that, if you had posted this anonymously, this question would have been closed in five minutes."

Maybe Felipe was right. It should be noted however that Bill Thurston posed his question in a really thoughtful and reflective way, drawing on decades of his own deep experience, and already had high "credentials" for asking such a question (being the author of On Proof and Progress in Mathematics, among other things). So as soft questions go, his I think was rather high quality. It got a pass.

I think that over the course of the existence of MO, there has been a trend toward becoming more hard-nosed toward "soft questions" -- this could be due partly to the rise of new users since the move to the SE platform. Is there some sort of collective sense that new users should "prove themselves" first, in the form of solid mathematical questions that indicate their serious engagement with mathematics and MO, before they have "earned a right" to ask softer questions? I really don't know (and don't know whether I have framed the question well, or tendentiously).

Perhaps the discussion should look toward more recent examples, where someone (of high MO "rep" or IRL rep) has asked a question that some like Peter would argue that those of lesser "rep" couldn't get away with?

Edit: Another not-so-recent example of a respected mathematician posting a reasonably highly upvoted question (+33-14=19) where some users wondered whether the question was MO-appropriate or would be seen as such if posted by someone with less rep: Most intricate and most beautiful structures in mathematics.

• (+1) This is a good answer, and I appreciate you taking the time to consider the issue. Hey, I would be quite happy if MO had a rule that the "soft-question" tag was not available to those with reputation $<N$. Instead, we let people use the tag and then trash them (though 90% of the time correctly) for using it. – Peter Dukes Mar 27 '14 at 20:34
• I made this comment Community Wiki to encourage anyone to add cases they know about where a high rep user has asked a soft question which likely would have been closed if asked by a low rep user. – Todd Trimble Mar 28 '14 at 16:27
• @Peter: I don't think the software has the ability to restrict tags by reputation. Also, it seems that real-world reputation is more significant here than MO points. – Nate Eldredge Mar 31 '14 at 12:16
• re your addition: that question was even closed initially. Another example slightly more recent, but not much, of a question asked by somebody with high on and off site reputation that got closed initially mathoverflow.net/questions/62401/… (Note OP of the two is not the same while one could think so on a quick look.) – user9072 Apr 5 '14 at 8:27
• @quid True enough; so far I haven't produced any examples free from controversy. Can you come up with any? – Todd Trimble Apr 5 '14 at 12:58
• One might speculate that mathoverflow.net/questions/143339/what-is-the-amplituhedron would typically have been closed as dupe. But I am now somewhat unsure what precisely you are looking for; examples "for" or "against" the idea there is a bias. – user9072 Apr 5 '14 at 14:56
• @quid The point was to test OP's idea that there are clear cases of bias. The examples I produced didn't give strong support for this idea in my opinion, and neither did his. So I wanted to see what examples did support his idea. Your last example might suffice; thanks. – Todd Trimble Apr 5 '14 at 16:10
• Glad it is useful and thanks for the clarfication. This interpretation is what I thought after a second more careful reading. Yet initially (for my first comment) I just read the addition and had more the opposite context in mind, whence that example there and my incertitude. – user9072 Apr 5 '14 at 16:56

A quick answer, a little bit too long to be a comment. It's not directly about your larger point (that there is some biased toward not-closing high-rep users' question), but it is about the too examples of questions you start your discussion with. Independently of who ask them, it seems to me perfectly reasonable that the first question got closed fast while the others enjoyed a long and successful life.

Indeed, it is true that the first question doesn't concern mathematics or the mathematical work. It can be asked, and answered uniformly in all academia, from Economics to east-asian studies to literature, etc. Moreover, it is a political discussion.

But the second question is really about mathematics, specifically the process of refereeing research papers in mathematics. Make no mistake: it is true that the question does not mention explicitly mathematics, but this because it is understood on this site. (Simlarly, if I see asked whether your favorite group is compact here, I know this is not about the group of your friends on Facebook.) Consider for instance subquestion 3):

3) What to do about papers with major grammatical or syntactical errors, but that are otherwise correct? Does it matter if the author is clearly not a native English speaker?

Of course the answer will be very different according to the field. If a paper is submitted in english literature, that it will matter a lot if there are grammatical error -- this is not acceptable for a paper in this field. On the other hand, the answer will also be different for a field like biology or other experimental science, where most paper have very little english text and consist mainly in a telegraphic description of en experimental device and tables containing the results of the experiments. Math papers are different: they are a discourse telling a story. Style matters, even if not as much as in a literary paper.

Or consider:

5) What do you do when you do not understand an argument? Does it matter if it "feels" correct. How long should one spend trying to understand an argument? 6) What to do about papers that have no errors but whose exposition is hard to follow?

It is obvious that those question make sense only for mathematics, or closely related fields, where an argument is either correct or incorrect with no in-between, and where similarly one can either understand, or not, a proof. It is also a fundamental question for mathematicians, which, after some level of seniority, spends a non-trivial mount of their working time refereeing papers.

• Your point is generally true, though you must admit: all that matters is other disciplines' perception that they are reading a water-tight argument. Whether they are or not does not change a referee's job. Plus, I would not count my life on the correctness of (some) successfully refereed math papers. Also, I argue above that the issue of fighting with journal publishers (an issue raised in the downvoted post) is particularly relevant for mathematicians. – Peter Dukes Apr 5 '14 at 19:56