![](https://secure.gravatar.com/avatar/01aa7d6d4db83982a2f6dd363d0ee0f3.jpg?s=120&d=mm&r=g)
On Nov 28, 2019, at 07:50, Guido van Rossum <guido@python.org> wrote:
Everybody has long presumed we'd go with 3.10. Maybe we're not following semver to the letter, but this part of it we follow -- 4.0 would mean a major rewrite or incompatible change.
For a long time I had hoped that Larry Hastings' Gilectomy project would succeed, in which case that would be a logical candidate for 4.0, since it requires a lot of incompatible C API changes. But Victor seems to have a better plan for evolving the C API, and the Gilectomy doesn't seem to be proceeding.
I’ve always expected that 4.0 would be reserved for such incompatible C API changes. I remember we had a similar discussion when 2.7.x was rolling over to x > 9. We didn’t bump it to 2.8 then, so I think we should just be prepared for Python 3.10 when the time comes. -Barry