Having just submitted a paper on Friday (yay!) I thought I would write this week about the process of letting go. This paper has been a long time in the making (it is based on my Ph.D. work, and as some people have recently pointed out, epsilon is not all that small anymore). There were many reasons it took me so long to submit, but one of the main ones is that the paper still felt like it wasn’t done. This is why I say that we let go of papers, we don’t really finish them.
For accomplishing many things in life, it helps to have a deadline. I was thinking of submitting my paper to Journal X. Every week that I worked on it, I decided it wasn’t “finished”, and I would tell myself “It’s OK, I can just work on it for one more week and then submit”. If you’re not careful, this can span years of your life (like with yours truly). So give yourself a deadline. Just make a date up, pretend like you’re going to be fired from mathematics if you don’t finish (eventually you will be, right?). In my case, I decided to apply to a conference instead (with some encouragement from people at a conference I’m currently attending). Publishing in a conference proceedings volume is usually not as prestigious as in a journal, although of course that depends on the journal and the conference. In the case of computer science, for example, proceedings volumes are the normal place to publish your work. Since the work I did was computational and the conference is modeled after computer science conferences, people have told me this should be counted (if accepted) as a good publication. The advice I have gotten, with respect to presenting this publication (again, if accepted) to a personnel committee, is to add to my tenure dossier all of this information. You need to make your case for tenure, so explaining how Journal X or Conference Proceedings Y rank in your area is always a good idea. At any rate, if not accepted, the paper is much closer to finished than it was a week ago, and I will get a chance to incorporate the feedback from the referee.
Another thing that is good, in life and in math, is asking people for help. My Ph.D. advisor helped me a lot with the earlier drafts. I radically changed much of what was in the paper thanks to his comments, and I agree that it makes a lot more sense now. As I said earlier, this past week I have been at a conference, and one person here took it upon herself to be my cheerleader/proofreader, which made a huge difference. Just having someone read your writing and tell you about how to use the DeclareMathOperator command in LaTeX so your operators are not italicized is amazing, but even better is the fact that they are also telling you if a definition is not clear or whether you really need to write the proof of Lemma A. She definitely gets a big acknowledgement in the paper. But this is not the only way to get help. One thing I haven’t done but many people do is to post an almost finished version on the Arxiv. People do read these papers and give feedback sometimes, but it’s not a sure thing.
A big source of feedback is also the referee report, although that may be tainted by feelings of rejection (if the paper is rejected). But still, you cry about it for a day, and then you have some potentially useful information (unless you are told, like a friend of mine did recently, to enlist the help of a “native English speaker”, being one herself). In general, you will have a better idea of what is not clear, whether there are gaps in your arguments, and sometimes you might even find out you are wrong. These are all good things to know! Hopefully, instead of the paper being rejected, you get a “review and resubmit”, which is another step towards getting your paper to that “finished” form.
So now I open this up for discussion, since I’m sure there are wiser, more experienced mathematicians who could give some great advice. How do you motivate yourself to finish/submit a paper? Where and who do you send it to? Is there a process you follow that you find is efficient?
Where to publish what is something I’ve thought a lot about. Clearly different people have different opinions on the matter, and it depends on one’s goals (loosely speaking, my goal is getting tenure while maintaining sanity).
Sometimes I choose a journal based only on subject matter (J. Stat. Phys. publishes lots of random matrix theory paper, so if I have a nice but not huge result in this area, I will think about sending a paper there). Sometimes I choose a journal based on the editorial board; recently I wrote a paper on certain generalizations of determinants which appear in random matrix theory. This paper was highly specialized and technical, so I sent it to Monatshefte, since some of the editors are from Vienna where determinantal evaluations are especially appreciated. Indeed, this paper went to the right referees and was accepted.
I personally am not a huge fan of general journals, since one runs the risk of the paper getting to an editor (and therefore referee) who might not be closely aligned with the area of the paper and therefore might turn in a tepid or apathetic recommendation. (The worse referees seem to be those that are near enough in area to actually accept the job, but not so near that they can appreciate the math you’ve presented). This being said, general journals—at least some general journals—are seen as being more prestigious than subject specific journals. And, in my department at least, that seems to carry some weight. So with my bigger results I’ll often attempt a shot at a very good general journal, with the understanding that rejection is more probable.
One also has to balance the rate at which journals move. (I recently sent a paper to J. Phys. A, partially because I thought physicists might appreciate the result, but also because they have a spectacularly fast turn around rate; perhaps at the expense of rigor, but it’s a physics journal!). The rate at which certain career deadlines (applying for jobs, tenure, promotion) are approaching is important too. In my department “submitted” papers carry very little weight (though are indicative that someone is active) compared to “accepted” papers.
My adviser said something one day that really helped me get my first paper sent off. He reminded me that it is foolish to try to make my first paper the best thing I’ll ever write. In fact, it would be a shame if it were. If you progress as a scholar, the first paper should be the worst one!