You're reading the public-facing archive of the Category Theory Zulip server.
To join the server you need an invite. Anybody can get an invite by contacting Matteo Capucci at name dot surname at gmail dot com.
For all things related to this archive refer to the same person.
Joe Moeller (Sep 05 2026 at 02:00):When complaining about the inherent troubles of being interdisciplinary, I’ve more than once been told to write “two versions of the same paper”, each targeted at a specific audience. I’ve never done it, and never talked in depth with someone who has about it. Obviously I wouldn’t take the advice too literally, to the point of academic malpractice. But I’m writing a paper now that would sorta benefit from this type of chopping. I’d like to hear perspectives from people who have or have not done this.
Adittya Chaudhuri (Sep 05 2026 at 05:58):Hi Joe! I am currently collaborating with a cancer biologist in a cellular plasticity problem associated to cancer metastasis. Many times, nowadays I also think in a very similar direction as yours, and would love to know other people's perspective on it.
David Michael Roberts (Sep 05 2026 at 10:30):One of my PhD supervisors did this a bit over 20 years ago, with a mathematical paper and a theoretical physics paper, describing essentially the same result, but tailored to two audiences.
John Baez (Sep 05 2026 at 16:00):Although I recommended that Joe did, I've never exactly done it. However, I helped write 2 papers on a category-theoretic approach to stock-flow diagrams, and software based on this. One was short, for the applied category theory conference. The other was long, for public health scientists. The second included lots of code. But we'd also developed more software that works on a web browser. So it wasn't just a repeat of the first paper from a different viewpoint.
Then my coauthors wrote a third paper, an explanation of the basic principles of this sort of categorical modeling for nonmathematicians.
I should see if anyone has cited these papers, and where.
Ryan Wisnesky (Sep 07 2026 at 00:47):Might this refer to taking a paper rejected at one venue and altering the framing and choice of results for a different venue? Say, math to CS or CS to engineering (the latter being something I'm engaged with right now)? I feel like that happens a lot more with my ACT papers than papers purely within a discipline (where framing/content would not be altered much by a venue change.)
Alonso Perez-Lona (Sep 07 2026 at 09:49):I am quite junior myself but I am in a similar situation right now on a topic in the middle of hep th and category theory. I originally aimed for a hybrid sort of paper but not only did the draft get very long, but most of the people from one of the two camps (hep-th) found the math too overwhelming even if the results themselves were interesting. I recently decided to split the draft into two and it felt somewhat liberating; it became way easier to spot what to highlight, and what to spend more or less time on
JR (Sep 07 2026 at 12:49):https://www.jmlr.org/papers/v11/carlsson10a.html
https://doi.org/10.1007/s10208-012-9141-9
Joe Moeller (Sep 09 2026 at 17:13):Ryan Wisnesky said:
Might this refer to taking a paper rejected at one venue and altering the framing and choice of results for a different venue? Say, math to CS or CS to engineering (the latter being something I'm engaged with right now)? I feel like that happens a lot more with my ACT papers than papers purely within a discipline (where framing/content would not be altered much by a venue change.)
It’s not what I meant. Maybe you were talking about someone else’s message.
Joe Moeller (Sep 09 2026 at 17:30):I’m in the same situation as @Alonso Perez-Lona , but with control theory and categories. I wrote like 30 pages of category theory, at the end I get a novel (I hope) result. But control theory papers have to be 6-8 pages, and they only tolerate minor upticks in math levels. So I’m considering two papers:
John Baez (Sep 09 2026 at 21:50):Sounds good to me! This lets you write to two quite different audiences.
Oisín FC (Sep 23 2026 at 15:41):I have similar question to Joe. Is there a standard policy with submission to theory CS conferences and pure maths journals? We would rewrite the introduction and significantly cut/change preliminaries between versions but probably not change the proofs or exposition around the main results. Would this run afoul of the double publication rules of maths journals? Essentially, the reason is, I like journals and my coauthors like conferences, so it means everyone is happy.
John Baez (Sep 23 2026 at 15:48):I would never include the same proofs in two publications - that seems illegitimate. Instead, what I tend to do is prove things in a math paper and use the results in another more applied paper, citing the math paper.
Oisín FC (Sep 23 2026 at 15:55):John Baez said:
I would never include the same proofs in two publications - that seems illegitimate. Instead, what I tend to do is prove things in a math paper and use the results in another more applied paper, citing the math paper.
People seem to write conference papers and then follow up with a longer journal version quite frequently, see:
https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.FSCD.2019.24
https://lmcs.episciences.org/10064
But usually the journal version appears in very specific venues.
John Baez (Sep 23 2026 at 15:58):People seem to write conference papers and then follow up with a longer journal version quite frequently, see.
Computer scientists, maybe. Not mathematicians, I think.
Do they include the same proof in both papers? That seems like a waste of bits: you can just link to the paper that has the proof.
Joe Moeller (Sep 23 2026 at 17:19):Right, I’ve seen this a bunch in engineering too. The more cynical actors have suggested I write a journal paper and then extract a handful of conference papers before submitting it. Perhaps this is viewed as legitimate to their field, but I can’t justify it to myself.
Joe Moeller (Sep 23 2026 at 17:21):I’ve also heard (this is very bad) of people plagiarizing someone and then claiming the original author actually plagiarized them, because the plagiarist put out their paper after the original author’s conference paper, but before their journal paper.
John Baez (Sep 23 2026 at 17:43):Joe Moeller said:
Right, I’ve seen this a bunch in engineering too. The more cynical actors have suggested I write a journal paper and then extract a handful of conference papers before submitting it. Perhaps this is viewed as legitimate to their field, but I can’t justify it to myself.
If this is truly legitimate in their field - as opposed to cynical actors merely claiming so - then I think it's okay for you to do this, at least if you're 1) publishing in journals in that field and 2) still looking for a job.
But I don't think it's considered legitimate in math.
I (again) think it's okay, even good, to publish a math paper with proofs of abstract theorems and a short discussion of applications, along with an engineering paper that applies those theorems and cites the math papers. But this is different.
Ryan Wisnesky (Sep 23 2026 at 18:48):It was not uncommon, at least a few decades ago, for CS conferences to have strict page limits but meet frequently, whereas journals had few page limits but only published infrequently. So it was not uncommon for a result to appear at a conference with the theorems stated without proof, and then an "extended version" to appear in a journal later, fleshing out the proofs and adding more content. The papers would have the same title and literally say "extended version", and the journals would even say, "must have 30% new content" or whatever. Sometimes the conferences would even say that there would be a follow up journal for extended editions.
John Baez (Sep 26 2026 at 19:37):That's perfectly reasonable, and it's different from publishing the same proof in two different venues.