11

Sorry if something like this has already been asked, I searched but I couldn't find anything similar to my question.

I'm a senior undergraduate and currently doing my senior thesis. My senior thesis is not original work, however it's quite demanding and I'm learning a lot of high level topics. I have been lurking around arxiv and started reading "Solved and unsolved problems in Number theory" by Daniel Shanks. My plan is to work on some open problems and play around with them so that I can try to get a publication before I graduate. My main reason for trying to get a publication is to increase my chances to get into a good graduate program (my GPA is not that great and I don't have the money to apply to many programs, so unless I publish something I'll probably only apply to safety schools).

With that being said if I were to do original work, how would I go about publishing? I might end up modifying a problem too much and proving something that might not be interesting, so I feel it'll get rejected from a journal for not being profound. I will also attack problems with all I know, so I might also end up using some heavy tools that aren't part of an undergraduate curriculum so I don't if i would send them to an undergraduate research journal. Maybe I could just upload on dropbox or arxiv, but then it's not a publication.

I have thought about asking my advisors about this, but I'll rather not since I'm aware I'm probably being overly ambitious and should probably focus on my thesis instead. Which I can agree with, hence I'll probably play around with problems on the weekends only or once a week. I'm also aware I might end up not publishing anything all, however in my mind unless I give it a shot I won't know. Either way I'll have fun and end up learning a lot about research so I don't see a downside.

(In case my background is relevant, my senior thesis is about perfectoid spaces. I've already taken a graduate course on commutative algebra, have taken a basic course on p-adic analysis,started learning about point free topology, already know the basics of category theory, still learning more about algebraic geometry, will learn about adic spaces soon/already know a bit about krull valuations, learning about homological algebra through weibel's book, started reading szamuely's galois theory book, will have to learn about etale cohomology soon, will also learn some things from almost mathematics, etc.)

David White
  • 29,779
Javier Ruiz R.
  • 213
  • 28
    Talk to your advisor about it. (This advice works for almost all situations to which it applies.) – Sam Hopkins Jul 16 '20 at 19:27
  • 3
    I think this is a good question, but unfortunately more suitable for academia.SE since the same question could be asked in about any other field. – leo monsaingeon Jul 16 '20 at 19:38
  • I'll try to muster the courage to send them a message then. – Javier Ruiz R. Jul 16 '20 at 19:39
  • My bad. I'm sorry if I wrongly posted here. – Javier Ruiz R. Jul 16 '20 at 19:41
  • 4
    If I am inferring correctly that you're trying to write a novel paper about perfectoid spaces, then I admire your courage but this is a tough task indeed. – Stanley Yao Xiao Jul 16 '20 at 19:41
  • My intention is not to try to write something that has to do specifically with perfectoid spaces. My intention is to try to read some papers and to read that book i mentioned, so I can play around with problems to see how I can try to apply what I know to prove something. – Javier Ruiz R. Jul 16 '20 at 19:45
  • It is always a good idea to have problems to think about and to motivate you to study. However, why are you in a rush to think where to publish before you even started to work on anything? If and when you'll have something concrete, then you should start thinking about it. (And as mentioned above, talk to your advisor.) – Yiftach Barnea Jul 16 '20 at 20:00
  • 7
    You should also be aware that getting a manuscript accepted at a math journal is often a lengthy process and it's quite common to have to wait upwards of six months to hear back from referees. Depending on the timeline for your applications, take into consideration that even in the best case scenario where you are able to write up a novel result, you might only have, say, an arxiv preprint and a journal submission at the time of the application. – Yoav Kallus Jul 16 '20 at 20:13
  • @ Yiftach My question was basically to know beforehand if I could even publish something even if it's not profound or interesting . But yeah I'll talk to my advisors. – Javier Ruiz R. Jul 16 '20 at 20:14
  • 1
    Starting your research with a known unsolved problem? ... That is probably a sure way to fail. – Gerald Edgar Jul 16 '20 at 20:39
  • 2
    @JavierRuizR.: Concerning your previous comment: What is "profound" and what is "interesting" heavily depends on the beholder. Anyway, I think the minimum requirement to (try to) publish a result is that at least in your eyes the result is interesting. Hence, the question "if I could [...] publish something even if it's not [...] interesting" sounds a bit... odd to me. – Jochen Glueck Jul 16 '20 at 20:39
  • I think as a rule of thumb maybe 50% at most of people's MSc theses (in physics) get whipped into shape and used in journal articles, although it might be less, I'm not sure. In the UK at least it helps if you go for an area of study where the University is renowned as it's then more likely that you will get in with the right supervisor and come up with something which can be published. – Hollis Williams Jul 17 '20 at 12:29
  • 1
    Why would you limit the schools to which you apply, with or without publication? If the issue is monetary, I think most reputable schools have mechanisms to request a fee waiver. – LSpice Jul 19 '20 at 16:27

2 Answers2

23

There are many undergraduate journals that would not be likely to reject your work as "not profound enough." Basically, if it's written while the author was an undergraduate, and contains anything novel at all (at the level one would expect of an undergraduate), then a journal can be found for it. This includes well-written expository accounts of existing texts, especially if they work out some more examples. The most famous undergrad math journal is Involve, which has higher standards than the others (more like what a grad student or professor might do). But there are plenty of others that accept papers that might not make it to Involve:

Like the rest of us, you should do the research first, and think later about where it can be published. I agree with Sam that this should be guided by your advisor. Don't be afraid to have a conversation with your advisor stating that you are hoping for a journal paper in one of these journals. That can guide how focused the research experience should be, and can guide how you write the thesis. It's not overly ambitious at all.

One last point: I don't think it makes sense to tether your perceptions of admission to grad school to whether or not you have a journal publication. For one thing, even after you finish the research, it'll take you weeks or months to write the paper. Then, there will be 6-12 months while the paper is being refereed. Then a back and forth with the referees. The point is: grad programs in math would be crazy to expect undergrads to already have publications before applying. However, if you have a good draft, you can share that when you apply. I think it would matter less than your recommendation letters (another reason to talk to your advisor often), GPA, and test scores. Having fun with it, as you say, is the best idea. Whether or not you prove new results, your advisor can relay your passion and depth of understanding to graduate programs, and that will matter much more than your publication record at this stage of your career. Good luck!

David White
  • 29,779
  • 1
    Thank you for all the info and advice! – Javier Ruiz R. Jul 16 '20 at 20:19
  • Yes agreed, having publications in journals is certainly desirable when applying to grad school, but in no way expected. The main thing you need to get onto a good PhD (in the UK anyway) is high marks on your exams. – Hollis Williams Jul 17 '20 at 12:31
  • Can you please give me some information about same thing but for grad student?....that would be helpful for me – annie_lee Jul 18 '20 at 15:13
  • @annie_lee Great question. I'll just write another answer below, rather than try to cram it into a comment. – David White Jul 19 '20 at 15:49
  • @DavidWhite thanks... That would be helpful for me also... – annie_lee Jul 19 '20 at 15:50
  • @annie_lee: Done! – David White Jul 19 '20 at 16:08
  • @DavidWhite: I came across this question through a google search, and found your answer to be very helpful. I am interested, though, in finding out more about Involve. Do you happen to know what the acceptance rate is (approximately)? Is it only the most talented undergrads that get published there? – Anonymous Oct 28 '20 at 20:52
  • That's a great question, and one that, unfortunately, I don't know the answer to. They do seem to publish a lot of papers (e.g., in 2019 looks like maybe 96 or so), so don't be afraid to try. If you need to know the acceptance rate, I'd recommend checking out the editorial board (their research interests are also listed) and asking an editor in your area, or the editor in chief. Good luck! – David White Oct 28 '20 at 22:51
5

I'm adding this answer in response to Annie Lee's question, because it's too long to fit in a comment.

Publishing as a grad student should definitely be done in consultation with an advisor. Unlike publishing as an undergraduate, a grad student's first papers serve as their introduction to the experts in their field, and largely determine whether they get a good postdoc. For this reason, it's very wise to have papers on arxiv before hitting the job market, but in many fields it's ok if these papers haven't been published yet (as long as the advisor's letter certifies that the work is solid). It's important not to put out anything shoddy, because your reputation really matters at that stage of your career. There's also the danger of getting scooped, e.g., if your PhD thesis has two parts, and you post the first part to arxiv a full year before the second, someone else might come along and do the second part before you can. An advisor will help you determine how likely this is to happen, and the pros and cons of advertising your work before it's fully finished. That's an important conversation anyway, so you can determine how much to say at conferences. An advisor can also talk to other experts in the field to encourage them not to try to prove what you are trying to prove in your thesis.

All that said, I think it's very valuable for grad students to have an early introduction to the world of publishing, as I previously wrote here. If you've never submitted a paper, responded to a referee report, etc. this can be anxiety inducing to do during your first job, while trying to juggle a million other things. Also, having an actual publication shows postdocs that you have what it takes to see a project through to completion. I was fortunate to write a paper as a second year grad student, in a totally different area than my main research (homotopy theory), and this gave me both confidence to get me through the hard times during my PhD research, and experience in choosing a journal, choosing an editor, when to keep polishing and when to submit, etc. Pro tip: the time to submit is almost always sooner than you think, because some degree of polishing can be done after the referee report, and the referee will always have suggestions for things to change (so, sending something too perfect can lead the referee to suggest big changes instead of obvious small changes).

In order to help give grad students this important experience with early publications, I joined the editorial board of the Graduate Journal of Mathematics, as I wrote about here. If you work out some interesting result or new example, and publishing it separately won't harm your PhD thesis, please feel free to submit to our journal! I think it's the only one of its kind, unlike the plethora of undergrad journals I mentioned in my other answer. To submit, you just pick an editor on the editorial board and email the pdf.

David White
  • 29,779