[Python-Dev] Move selected documentation repos to PSF BitBucket account?

[Brett Cannon]

On Sun Nov 23 2014 at 1:31:36 PM Brett Cannon <brett at python.org> wrote:

> On Sun Nov 23 2014 at 11:56:49 AM Guido van Rossum <guido at python.org>
> wrote:
>
>> On Sat, Nov 22, 2014 at 10:49 PM, Nick Coghlan <ncoghlan at gmail.com>
>> wrote:
>>
>>> More generally, I'm very, very disappointed to see folks so willing to
>>> abandon fellow community members for the sake of following the crowd.
>>> Perhaps we should all just abandon Python and learn Ruby or JavaScript
>>> because they're better at getting press in Silicon Valley?
>>
>>
>> That's a really low blow, Nick.
>>
>> I think these are the facts:
>>
>> - Hg/Git are equivalent in functionality (at least to the extent that the
>> difference can't be used to force a decision), and ditto for
>> BitBucket/GitHub, with one crucial exception (see below)
>>
>> - We're currently using Hg for most projects under the PSF umbrella
>> (however, there's https://github.com/python/pythondotorg)
>>
>> - Moving from Hg to Git is a fair amount of one-time work (converting
>> repos) and is inconvenient to core devs who aren't already used to Git
>> (learning a new workflow)
>>
>> - Most newer third-party projects are already on GitHub
>>
>> - GitHub is way more popular than BitBucket and slated for long-term
>> success
>>
>> But here's the kicker for me: **A DVCS repo is a social network, so it
>> matters in a functional way what everyone else is using.**
>>
>> So I give you that if you want a quick move into the modern world, while
>> keeping the older generation of core devs happy (not counting myself :-),
>> BitBucket has the lowest cost of entry. But I strongly believe that if we
>> want to do the right thing for the long term, we should switch to GitHub. I
>> promise you that once the pain of the switch is over you will feel much
>> better about it. I am also convinced that we'll get more contributions this
>> way.
>>
>> Note: I am not (yet) proposing we switch CPython itself. Switching it
>> would be a lot of work, and it is specifically out of scope for this
>> discussion.
>>
>
> If we want to test the complexity of moving something to GitHub then
> probably the best repo to use is the peps one:
>
>    - Very few people directly use that repo (you and me alone could
>    probably manage it if we enforced all changes through a PR as I could then
>    do approvals from work instead of having to wait until I was at home with
>    an hg checkout available)
>    - It's used on the website so it would require updating infrastructure
>    - It isn't a lot of overhead to tell people who email the peps mailing
>    list to "please send a pull request through GitHub" since it isn't tracked
>    in the issue tracker anyway
>    - There is a benefit of setting up some CI integration to know when a
>    PR is actually incorrectly formatted
>
> And if people want to test the impact of Bitbucket we could do it for
> something like the HOWTOs as that too involves infrastructure but is not
> used by a lot of people. In fact we can make it known we are piloting this
> approach on Bitbucket and see what kind of contributions it triggers (ditto
> for the peps since I'm sure some people will want to send in typo PRs and
> such).
>

Actually the tutorial might be best to measure ease of contribution for
people on Bitbucket since we can also ask people who use the tutorial to
test out pointing people to the Bitbucket repo if people want to send a PR.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-dev/attachments/20141123/1985046c/attachment-0001.html>