[Python-Dev] pep42 procedure?

[Tim Peters]

[Mark Hammond]
> That would be a great idea, if only we had a PEP author.  To be honest,
> I really dont care too much about this issue, and have been doing my
> level best to avoid it.  But Tim knows that, and delights in assigning
> related patches to me ;-)  I certainly don't care enough to own a pep
> on it. ...

I assign them to you because there's barely a line of Windows startup code
you didn't write, and you gave Windows the notion of naming the main DLL
differently across (some) different releases thus creating the possibility
for the problem these patches are trying to solve <wink>.  That's not to say
it's "your problem", but is to say you've thought more about "stuff like
this" than anyone else, and have very particular ideas about how you want
"stuff like this" done:  you're the best qualified.

That said, the subject of the PEP shouldn't be competing hacks to worm
around Windows-specific startup problems, it should be an actual design for
binary compatibility across iterations of the C API, or at least a reliable
way for extensions to *know* they're not getting a version of the Python
core they can use.

we're-only-hacking-because-we've-ignored-the-real-problem-ly y'rs  - tim