Why is the Migration to 3 Taking So Long?

(The conclusion is kinda obvious, if you are in the field long enough.)

@juliobiason Has Python officially adopted, "Learn nothing from Perl" as their motto yet?

@tfb Not sure if that's the same thing.

Perl 6 tried to change the language itself, while Python 3 changed an underlying property (Strings being unicode) that forced the major rewrite.

Sure, the results can be comparable, but the reasoning behind each is completely different.

@juliobiason Oh sure, I agree. But Perl learned the lesson and decided to properly maintain P5 to avoid creating a disaster, whereas Python seems determined to getting to the disaster.

@tfb @juliobiason The disaster was when you couldn't move to py3 because your favorite pypi package hadn't.

That's behind us now, ever since py3.3 finally made it possible for a package to have a single py2/3 codebase without any translation shenanigans, and packages slowly caught on. Some PSF-paid porting helped.

People with legacy codebases will be able to run legacy python code with legacy python maintained by distros, and PyPy.
@juliobiason @tfb The lesson Python should learn (but hasn't, as async suddenly became an illegal module name in a point release) is to introduce incompatible syntax with `from __future__` first, and then make it default in the next major release.
Sign in to participate in the conversation
Functional Café

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!