Summarizing what I've seen come by here: - the examples we have of bits of software developed as part of sage that ended up as library components of other projects are peripheral, interfacing parts that were spun off into independent libraries.
- we don't have examples of core functionality of sage getting adopted as a component of another project - licensing under GPL may or may not be an obstruction to this. - relicensing core parts of sage may well be impossible because it would involve tracking down all relevant contributors to agree to relicensing and that may be infeasible (plus, each individual developer on that list would have a veto). Against that background: what purpose would modularizing the core of sage serve and what benefit do we think to derive from it? It's a nice idea that sagemath code would be reused elsewhere, but against the backdrop that we have here I'm not so sure it's realistic to expect this (and I'd wonder what benefit *sage* would have from this). Note that the message quoted at the start of this thread: https://lists.debian.org/debian-science/2024/07/msg00023.html wasn't about code reuse at all: it was from a debian packager who took source distributions from pypi to build "manageable" chunks. Quite frankly, I don't see why a pypi package is more manageable if it is split up (wouldn't one monolithic pypi distribution work just as well?), but if this splitting up makes packaging sagemath more palatable for linux distributions then I think we have a real win. Once distributions are happy packaging sage in a conventional way, perhaps people could easily generate flatpaks or snap packages as well, which should be more stable (they'll be huge, though). I could see how there are other arguments for splitting sage up: components like jupyter and much of the docbuilding infrastructure is only very weakly tied to the sagemath codebase: communication takes place through protocols and file formats rather than through an API. So packaging sagelib without these and then perhaps a distribution that bundles sagelib with a jupyter server for convenience could make things easier (but it might also be that we're already very close to doing that anyway, and perhaps we literally do that) -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/feff98e8-b5ef-4a37-93a4-ff664d216c91n%40googlegroups.com.
