I am unsettled by this post which concerns untimely death: http://mathoverflow.net/questions/48477/never-appeared-forthcoming-papers/53164#53164 . However, I do not want to argue for its removal on the basis of it not properly answering the question. I want to stimulate discussion towards a particular end: deciding if such material is appropriate to appear on MathOverflow?

The more cogent parts of my initial reaction to this post were on the order of revulsion, the thought that this could trigger copycat episodes, the idea that salacious bits of gossip make for tempting reading, and other things which I find socially negative in the context of (my idea of) the MathOverflow community.

I can argue on the other side that real life happens, that tragic affairs are not banned from the mathematical culture at large, and that a more mature and hardened soul might acknowledge Gerry Myerson's post and move on. This argument still leaves me unsettled.

What I want is to be settled, either by community agreement that the post is inappropriate for MathOverflow proper and thus should be removed, or a discussion by enough of the community to come up with a rationale for why it is appropriate, or a policy guideline from an administrator that says why such material is to be allowed. I don't expect to have the community serve my whim, but there is an issue raised by this post, even if I can't clarify what the issue is. I think it important enough to discuss and see if this is a pants-wearing issue, or something even more serious.

I want to make clear that in other contexts, I would not object and might even welcome the information in the post (although not for enjoyment or entertainment). Also, despite my reaction above, I believe Gerry Myerson is correctly quoting and attributing his source. My concern is that this bit of information is misplaced and that it is to the detriment of MathOverflow. I wish I could clearly explain why, and hope that someone else will.

Gerhard Paseman, 2011.01.24

I have no problem with the post.

I'm with Mariano and Andres here. I see nothing inappropriate in the post.

I appreciate the responses thus far, even though I am still concerned about whether the post is appropriate. I wanted to make clear again (since Gerry's response suggests that I did not) that I did not consider the content itself to be gossip. I am still having trouble clarifying my reaction, but I understand that what I feel is not necessarily what is. Again, I believe that Gerry Myerson is accurately quoting a source, and is not relaying gossip, regardless of how I feel.

Also, I still think the issue involved may not just a matter of wearing pants. I would appreciate hearing from a few more members before letting the matter rest. I will further work on clarifying why I think it is inappropriate.

Gerhard Paseman, 2011.01.24

A short lived theory!

It may be something with the accents in our names...

You are the best. Ever.

Coincidentally, I heard this story independently last Wednesday while visiting another university. However, it was told in hushed conspiratorial tones and without any identifying details, so it felt much more like an urban legend. While I was a bit uncomfortable reading the post under discussion, upon reflection I think I'm happier knowing this is a factual story that I can put out of my mind, rather than the "salacious gossip" described above.

Karel DeLeeuw died pretty much the same way in Stanford, so the story is by no means unusual.

fedja, Halmos writes of Koppelman, "several graduate students had started to work with him, but none had ever finished." de Leeuw was very much the opposite; at Stanford, he had the reputation as the one who would get you through if you were having trouble working with someone else. I know this as I was a grad student at Stanford (though not a student of de Leeuw) some years before the tragedy.

Whilst I have no particular problem with the post in question, I find that I do share something of Gerhard's misgivings. It's hard to put my finger on precisely what it is that makes me uncomfortable about having this post on MO, but I think that my disquiet is more that a question that leads to an answer like this is high on my list of "Unsuitable for MO" questions.

I notice with some dismay that this question got highlighted on the "Not Even Wrong" blog as an example of how even in the tight confines of MO mathematicians can still "let their hair down". That bothers me. I would much rather that the first thing that comes to someone's mind when they think of MO is that it has led to (for example) many papers, or some unsolved conjectures being solved, or similar stuff - basically, the "success stories of MO" thread here, rather than that it's where people can get talk about mathematical urban myths. Given that people will naturally gravitate to the slightly odd and bizarre, the only way to ensure this impression of MO is to be strict in keeping stuff like this out.

I can't say that I am entirely settled. I am getting over the emotional impact of the post, and I am glad to see (what I presume to be) calmer heads than mine taking up the discussion. In particular, I am gratified to see that I am not the only one bothered by the post.

While I could ask for removal of the post on other grounds, it might look to others (and possibly to me too) as asking others to cater to my sensitivities. I don't want that. I prefer to argue on the basis of the good of MathOverflow and its community.

MathOverflow is like many things and not like others. People would be happier for it to have more discussions, and other people prefer it to have less. Some want softer, less technical questions, others want to keep the technical and truth content at a high level. In particular, some want it to be like a professional seminar, with as much informal talking about the subject elsewhere, sometimes in comments, sometimes not.

If I were to meet Gerry Myerson in person at a seminar, and he were to relate to me the information in his post word for word, even while the speaker was speaking, I would be bothered by the information, but it might clear up some other questions for me, and I would not object to it as I have here. If he were the speaker, and had to make passing
reference to the unfortunate victim because the subject of the talk involved a continuation of the victim's interrupted work (e.g. "whose career was cut short by an act of violence") , I might wonder at the propriety of the reference but would appreciate it being brief and containing as little detail as possible.

If the seminar were held in honor of the unfortunate victim, I can then see how the information Gerry posted would be relevant in presenting at such a seminar. Under no other circumstances can I see the rationale for someone coming up, announcing this fact however tactfully, in the context of a professional seminar. In the context in which Gerry made the post, it seems unacceptably jarring.

As a milder example of holding to this standard of not bringing up unpleasantness, I wondered whether Richard Arenstorf (cf. http://mathoverflow.net/questions/52489/on-the-non-rigorous-calculations-of-the-trajectories-in-the-moon-landings ) had a son Gerhard. I decided not to pursue it on MathOverflow because satisfying my curiosity on that point was not worth the harm it might cause. (Gerhard Arenstorf was an IMO participant who died accidentally in Nashville while under 20.)

My concern isn't with Gerry's sensitivity or wisdom in selecting what to post, and I do not want to make this an issue about him. I think the raw content of the post is something that should be avoided on MathOverflow. I prefer (at this time) to not open up issues of whether it is better placed in a comment than in an answer, or whether talking about how mathematicians died is an appropriate question for MathOverflow.

Even if Gerry's post is deemed acceptable by the community, there may come similar posts which are deemed inappropriate purely on content, or perhaps I mean purely on context. What is said here may provide useful for those future cases. As I said in my initial discussion post, under other circumstances I might welcome the information Gerry
provided. I just think MathOverflow is not the place for it.

I appreciate the time others have taken to contribute. If my thinking on this gets any more clear, I will share it here.

Gerhard Paseman, 2011.01.27

I honestly do not see any need for such a policy.

A tactful description of facts need not be obscured: we are adults. In particular, Gerry's post very precisely gives a historical example of a paper which did not appear, and explains the reason---it is on-topic, and to the point. It is tragic that that happened, of course, but it did happen. Since neither Gerry turned his post into a gory description of the tragedy, which would have been completely our of place, nor the ensuing comments did anything remotely similar to that, I really do not see what the problem is.

Mathematics is a human activity, with all the good and all the bad that that implies.

A good four fifths of newspapers would have to be rot13ed...

What Mariano and Pete said.

I do not feel sufficiently involved in MO to voice an opinion, I only wanted to point out a potential confusion.

@Andrew Stacey: The answer discussed here was made in 'Never appeared forthcoming papers' not in 'Urban Legends', the post mentioned on 'Not Even Wrong.'

I guess if you view MathOverflow as a site for purely mathematical discussions (akin to a support site for, say, C++ or Microsoft Word), then all this subjective messy human stuff really is inappropriate. If that's what the people in charge want, then I suppose they need to be more strict on banning all the questions that wander off into more subjective territory. But I confess to enjoying quite a bit what working mathematicians have to say about both the solid math questions as well as the messy subjective stuff. And I like being able to go to one place to see both. I don't know about the rest of you, but I just don't have time to check out *and* contribute to more than one site like this. So although I can see why some of you feel that stories like the one about Koppelman should not be mixed together with discussions about how to prove the converse to the Bishop-Gromov inequality, I find that for me convenience trumps focus.

But I confess to enjoying quite a bit what working mathematicians have to say about both the solid math questions as well as the messy subjective stuff. And I like being able to go to one place to see both. I don't know about the rest of you, but I just don't have time to check out and contribute to more than one site like this.

Why should I have to wade through it all just because you like it? If it is separate, I can easily choose to ignore it whilst it really isn't any more difficult for you to check two sites than one.

I want MathOverflow to be part of my working life, not my goofing off life. I find that my working day splits in to three types of activity: focussed research, wide-gaze mathematics, and "goofing off". My struggle is always to maximise the focussed research, but sometimes I just need a few minutes "goofing off" to recharge. Now, MO can't be part of my focussed research - that's when I'm trying to answer questions that I hope no-one else has a clue how to answer! But it can be a part of my "wide-gaze" work and I find it extremely useful at that point. But if it starts creeping in to the "goofing off" part, then it loses any value that it might have otherwise had.

There are plenty of blogs, forums, and other places out there for the "messy" stuff, which don't have quite such as high profile as MO and so are easier to be more informal. If it doesn't fit on MO that doesn't make it a Bad Thing, just not in MO's back yard.

I'm also curious about the human side of mathematics. But there are plenty of places to learn about such matters. There aren't many places like MO. I'm afraid that "I like being able to go to one place to see both" just isn't even an argument. As the old advert had it: "Take two bottles of shampoo into the shower? I just RSS and go."

Mariano: I know that you didn't really mean '18+', but your underlying assumption was that everyone reading MO was as mature as you. Sadly, I doubt that very much. MO is a very public place. I recommend that we act accordingly. After all, wasn't that the argument we kept throwing at Harry in the Bad Old Days? That his actions here would echo down the corridors of academia for generations? Shouldn't we also think about that?

an_mo_user: Yeah, I got confused. But if I can get so confused, so can anyone else. And the fact that this question, which seems innocent enough, could provoke such an answer underlies my point about keeping MO for focussed research questions.

On the wider SE network, it is no longer possible to ask CW questions up front (they can be made CW by a moderator later). There's a reason for that: they want to keep the quality of the questions high and they found that CW questions generally didn't contribute much to that level of quality.

@Andrew Stacey : I hope you don't take the following as an attack, because it really isn't meant to be. However, I'm sure you are aware that your argument can be turned around. Namely, why should the rest of us not get to enjoy the odd soft community-wiki post just because you can't resist reading it and don't want the temptation to goof off?

Of course, most of the soft questions are lousy and don't generate interesting responses (and I've voted to close plenty of them because it is clear that they aren't going to be interesting). However, some of them do and I've learned a lot from them.

I think the answer to this is to live and let live. It's easy enough for you to add the "soft-questions" tag to your list of ignored tags...

I agree with Andy Putman and Mariano.

@Andrew Stacy: If the "goofing off", "subjective", "historical", "professional" type questions were all appropriately tagged then people who are uninterested in them can just ghost those tags and focus on mathematical questions. People like me have certain desires for what we'd like MO to be and people like you have different desires, but I don't see why MO can't satisfy both groups.

With @Andrew Stacey and others, I wasn't thrilled with the original question. My response is that I don't read most questions like this (except for the barely enough time it takes to skim past it in Google Reader). So I would never have come across the answer without @Gerhard Paseman bringing it up.

That said, I strongly oppose any sort of "keep everything happy" policy. Of course, we have rightly closed questions from depressed students who should seek support elsewhere. But the answer in question was respectful and accurate, and anyway I don't like the idea that everything must necessarily be happy for it to be appropriate. (Cue rant about antidepressants and advertisements for them.)

Firstly, I know I'm pretty far out on one extreme on this position so while I argue my case strongly here, I'm not so strict on MO itself (I think - do tell me if you think otherwise). But I do feel that if I didn't keep speaking up about this then the "happy medium" would move too far to the soft end. So I argue not so much as to hope to convince anyone as to remind you that there are people who take quite a hard line on "soft questions".

Andy's reverse argument doesn't work. I wish it did. Unfortunately, for it to work then the tagging system would have to be a lot more consistently used, and the "ignoring" system actually do what it says on the tin. I have some tags as "favourite" but it's almost useless for me. In fact, I tend only to use the tags when the question title and the first couple of lines don't tell me enough about the question - the reverse of what I would have to do for the separation by tags method to work.

If the "goofing off", "subjective", "historical", "professional" type questions were all appropriately tagged

and if I could speak Norwegian then I'd find life in Norway much less confusing. Unfortunately, the two events are about as likely so I don't see the point in playing "what if" games about idealised situations.

The wider point is that it is much easier to amalgamate two separate sites, say via a feed reader or even just having two tabs open in your browser (good grief, how much effort does that take?) than to separate one site into two streams.

To Mariano, may I remind you that it wasn't so much the answer itself that gave me cause for misgivings. I agree that that answer was written sensitively and carefully, but it could so easily have been done badly, and so easily have been offensive. That it wasn't is a happy (!) chance and I'm not sure that we'd be so lucky next time.

To Theo: I am certainly not trying to "keep everything happy". I am trying to "keep everything mathematical".

Deane: Thank you. That's actually all I'm asking for. If each "side" remembers that the other side is also present on the site and that their own preferences are not universally acknowledged as a truth, then each will moderate their behaviour and both will find that the site is big enough to accommodate them.

Andrew, I don't understand what you mean by "[you] are not so sure we'd be so lucky next time": when something distasteful, offensive, racist, sexist, or whatever else gets posted, we'll do the only sane thing, delete the post. What other option there is? What other thing is there to do to prepare for it? (honestly, I don't know what it is, I guess next time I'll recognize it when I see it :) ) Are you presenting a slippery-slope argument? Right now, It seems to be, the slope is zero!

Enough time has passed that I am less disturbed by the post that raised the issue.
Although not entirely settled, I have the impression that many (but not all) in
the community deem the post as acceptable MathOverflow content. I thank those
who read and participated in this thread. I remind people that the issue was
about whether the content was appropriate, and not about whether Gerry Myerson
delivered the content appropriately.

Andrew Stacey said:

'If each "side" remembers that the other side is also present on the site and that
their own preferences are not universally acknowledged as a truth, then each will
moderate their behaviour and both will find that the site is big enough to accommodate
them.'

I think this is an important point: we may be unable to "do no harm" with each
post that we make, but if we are conscious about what we post, then most anything
reasonable will be accepted; the rest can be discussed on meta.mathoverflow and
resolved in proper fashion.

Although I would like Gerry to accommodate me (and he actually offered a compromise),
even more I would like to be aware of what content be allowed on MathOverflow. I agree with
Andrew Stacey's suggestion of a "Seminar test", even though that might remove some
of the levity I have enjoyed on MathOverflow. Now after having some community
members speak on this issue, I have more of an idea of the range of things that
the commmunity will (in general? on average?, certainly not unanimously) accept as
MathOverflow content, even if it doesn't meet my ideal of what should be on MathOverflow,
nor have all of it pass a "Seminar test". Having that "heads up" is worth something to me,
and perhaps that range should be formalized and added into published policy.

Finally I thank Gerry Myerson. With one exception, I have seen him behave
faultlessly on MathOverflow (and the exception isn't an issue of behaviour so much as of
what is considered appropriate content for MathOverflow, on which the community is still
not unanimous), and I look forward to his continued participation and contributions.

Gerhard "Getting Off The Soapbox Now" Paseman, 2011.03.02

I agree with Todd that Krantz's book is a strange reference. The book is indeed mostly fun and gossip and those who are familiar with the book might view the story relaed in that context. Having a reputable newspaper as a source for the story would have been much better. The question sets a bad context too. I don't think relating the stroy in the level of detail given would be inappropriate in every possible MO context though.

Yet, already, somebody vote to reopen...