Zulip Chat Archive
Stream: Formal conjectures
Topic: Referencing OEIS
Daniel Chin (Jan 30 2026 at 13:50):
Quick question regarding how to reference OEIS. From (https://github.com/search?q=repo%3Agoogle-deepmind%2Fformal-conjectures%20oeis&type=code), it looks like people are referencing OEIS with the following formats:
"[OEIS A6697]"
"[oeis.org/A067720]"
"[A34693]"
Do we have a standard way that we should be referencing these? Maybe a linter can enforce this?
Moritz Firsching (Jan 30 2026 at 21:40):
I'd say the preffered and probably most used format s [A228828](https://oeis.org/A228828). We still haven't decided on how to best handle references and at the moment it is a bit adhoc, but for OEIS references I'd recommend going with this:
Daniel Chin (Jan 31 2026 at 02:32):
Thanks! Tangentially, what's the proper way of naming multiple parts or variants?
For multiple parts, I've seen examples of ".parts" and ".part". Not sure if it should be plural or not. And for the individual part names, I've seen ".parts.a" and ".parts.i". Not sure if we should be referencing them as letters or roman numerals.
For variants, I've also seen ".variants" and ".variant".
Yaël Dillies (Jan 31 2026 at 08:14):
We should replace the namespaces in the variants' names with underscores, because eg variants.i.i gets flagged by the duplicated namespace linter
Last updated: Feb 28 2026 at 14:05 UTC