Open mfisher87 opened 2 months ago
Do we have an official policy?
We do not.
Also, are we following semantic versioning?
It's buried in the release guide, but we do!
We have a v2 branch in progress, and I'm fairly sure dropping support for a Python version justifies a major bump if we are. At least that's what I've seen in practice!
That makes sense and seems like a good opportunity to do this bit of housekeeping.
Thanks, Jessica! Does the new issue title seem like accurate and complete?
Hey all, I noticed in a previous post NEP 29 was used justification for dropping Python 3.6. It's been a couple years, and the community (including numpy) is now coalescing around this very similar standard, SPEC0: https://scientific-python.org/specs/spec-0000/
Under SPEC0 and NEP29, we can drop support for 3.7, 3.8, and 3.9. Do we have an official policy?
Also, are we following semantic versioning? We have a v2 branch in progress, and I'm fairly sure dropping support for a Python version justifies a major bump if we are. At least that's what I've seen in practice!