Bug#551926: [Python-modules-team] Bug#551926: Bug#551926: cannot be installed together with pip or pip-python

Sandro Tosi morph at debian.org
Sat May 8 15:43:11 UTC 2010


Hi Gregor,

On Sat, May 8, 2010 at 14:55, gregor herrmann <gregoa at debian.org> wrote:
> On Sat, 08 May 2010 08:27:40 -0400, Jonathan Yu wrote:
>
>> I see no reason for Perl's pip to have to change its name, simply
>> because the author of Python's pip chose a name which was already in
>> use by someone else, and because the author was already informed that
>> something like this might happen, and chose to proceed anyway.
>
> The situation as I see it:
> * Both upstream authors want to keep the name.
> * Sandro wants the python package to keep pip, Jonathan wants the

just to clarify that it's not that I want it (i don't use it, after
all) but given how strongly it's advertised (and by such a big player
in the Python community as Raymond Hettinger, that mentioned it also
today), then I see definitely a problem in renaming it.

> pip doesn't have any rdepends.
> I was curious what it was needed for in the first place; I had a
> hunch about padre, and indeed:
...
> And indeed, the graphs on http://qa.debian.org/popcon-graph.php?packages=pip
> look much like the dates when pip was added/removed as a dependency
> of padre.
>
> I'm not sure how useful/needed/... pip is on its own. But looking at
> the (non-existant) rdepends and the popcon values I think RMing it
> would be a viable solution.

also looking at python-pip pocon (260, ~5 times more than
pip-after-padre effect, ~20 times more than before) and how low is
pip, and that also one of its maintainers is challanging its
usefulness :) I think that maybe removing it would be the lowest
impact action.

Regards,
-- 
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi





More information about the pkg-perl-maintainers mailing list