News for hackers. Stuff that matters.
Re:Towards a New SymPy by bmitc on Friday September 08, 2023 @11:53PM (#37442128)
Mathlib is not nearly as complete as advertised. It is very much a collection of research projects with little cohesion.
Re:Towards a New SymPy by staunton on Saturday September 09, 2023 @08:24AM (#37444501)
A CAS also covers only a tiny part of "an undergrad math degree" so that's beside the point here. Assuming it had appropriate tactics doing the job of a CAS using all theorems (which isn't happening any time soon), what math would be missing?
Re:Towards a New SymPy by bmitc on Saturday September 09, 2023 @10:41AM (#37445986)
I'm not really sure what you're asking. I wasn't commenting on any of the CAS asoects of Lean or Mathlib.

> what math would be missing

Just generally speaking, manifolds. There's no clear or standard approach in Mathlib. There seem to be approaches, but they definitely aren't turnkey.