On 11 May 2014 09:50, Nick Coghlan <ncogh...@gmail.com> wrote: > Let me be clear: this is *not* a technical decision from my perspective. It > is a relationship management one, specifically in regards to maintaining the > still fragile delegation of authority from python-dev to PyPA.
Nick, Given that the current behaviour is directly mandated by PEP 438, it seems to me that this is even more so a decision about the authority of approved PEPs. If the decision is that pip implement any behaviour that contradicts the wording in PEP 438, that to my mind calls into question whether packaging PEPs are authoritative documents at all. If changes are needed to resolve Stefan's and MAL's concerns, doing so via changes to PEP 438 would be a much better approach, as the PyPA have no special voice in such a debate. That's essentially the whole purpose of the packaging PEP process. Paul PS To my knowledge, neither distil nor easy_install implement PEP 438, so I understand that the distinction between pip and the PEP is somewhat obscure. But that's not our fault - feel free to complain to the maintainers of those installers about PEP compliance if you feel it'd help... _______________________________________________ Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
