It's worth mentioning that in virtualenvs and conda envs, where there can only be one version of Python installed, 'python' refers to that whether it is Python 3 or 2. So it's already not a safe assumption that 'python' always means Python 2, even if you discount Arch.
On 15 April 2015 at 21:04, Scott Kitterman <skl...@kitterman.com> wrote: > On April 15, 2015 8:00:22 PM EDT, Barry Warsaw <ba...@debian.org> wrote: > >On Apr 15, 2015, at 04:42 PM, Scott Kitterman wrote: > > > >>Then I don't understand why the whole s/python/python2// plan in the > >shebangs > >>helps anything. As long as both exist, it's a no-op. > > > >Partly this is to begin to educate users to stop using /usr/bin/python, > >which > >has unclear semantics across the wider Python community. If users see > >distro > >installed scripts use /usr/bin/python2, and PEP 394 says to use it, > >they will > >switch over and in time (e.g. by 2020) the impact of removing > >/usr/bin/python > >will be greatly lessened. > > > >>P.S. It would be nice if there would be a PEP that says to never ever > >do > >>this. I know it would make Arch have a sad, but they'll get over it. > > > >PEP 394 is the vehicle for this, and getting the Debian and Fedora > >ecosystems > >aligned will be a powerful force for making sure it says what we want > >it to > >say. > > PEP-394 is very weak in my opinion. All it says is we aren't ready to > break existing systems yet, but we probably will in the future. I think > it's better not to do that period. > > Scott K > > > > -- > To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact > listmas...@lists.debian.org > Archive: > https://lists.debian.org/403c6d4f-1e52-4e67-aa6a-5914d0be8...@kitterman.com > >
