Akuli

Instead of generating new stubs, it would be better to move existing stubs to typeshed, if that's possible. And again, it doesn't matter if the stubs are outdated or incomplete...

There are quite a few `Callable[blah, Any]`s, but not really too many to go through them all by hand: ``` akuli@akuli-desktop:~/typeshed$ git grep 'Callable\[\[.*\], Any\]' | grep -v @python2 |...

I don't really like the `Void` idea. Knowing that `Void` exists, it would be weird to see `-> None` instead of `-> Void`. Many other languages use `void` as the...

The docutils stubs were added in #5192, and lots of things are missing from them, including the whole `Element` class. Feel free to create pull requests that improve them. Let...

`stubgen` is a tool that generates stubs from source code, but before we rewrite the stubs with it, maybe @srittau should tell us what exactly he did in #5192. Afaik...

Maybe the published list should go to a GitHub issue on this repo. Issues can be edited from github actions, and it would be very discoverable for contributors, especially if...

>str in py2 but bytes in py3 In this case, you should just use `bytes`. In Python 2, `str` and `bytes` are the same, and `unicode` is different.

PRs to improve the stubs are welcome :)

I like this approach: it's practical and correct enough. Here are a couple examples of how we have done this in the past: - If the stubs would have to...

When "balancing", we usually consider false negatives (no errors for wrong code) less bad than false positives (errors for correct code). >So making headers/cookies a Mapping in Session.request but MutableMapping...