[Python-checkins] r72563 - peps/trunk/pep-0374.txt

[David Goodger]

On Fri, May 15, 2009 at 11:22, Dirkjan Ochtman <dirkjan at ochtman.nl> wrote:
> Thanks for letting me know about your thinking on this.
>
> I'd like to state that Martin v. Lowis proposed this as a good way
> forward (updating the current PEP).

In this matter, I disagree with Martin. The original objective of PEP
374 *has* been achieved: a DVCS was chosen. PEP 374 should continue to
document that process. The transition details should be *added* to the
record, not replace it.

BTW, I can't find anything prior to Martin's reply to this thread, but
your message implies there was something before revision 72563. Do you
have a reference?

Have you seen the other replies to this thread? Others agree that this
revision should be reverted.

Question for you: what harm would there be in reverting revision 72563
and creating a new PEP?

> I know that changing the authors
> seems a bit strong, but all of their work has been removed from the
> PEP and moved to a separate wiki page, which the current version of
> the PEP clearly points to. The PEP, along with the contents of the
> wiki page, still clearly show who the authors of that work were. So I
> don't feel I've misrepresented the work that was done,

I don't think you've intentionally misrepresented anything or anyone.
Your intentions seem good. It's just the execution that needs fixing.

> though I'd be
> happy to restore the authors and/or the PEP contents if that's deemed
> more appropriate.

Good, thanks. I will leave that decision to the other PEP editors.

> I just followed the path that Martin recommended
> (since I don't know enough about PEP processes to have a strong
> opinion on this).

>From PEP 1, PEP Purpose and Guidelines
(http://www.python.org/dev/peps/pep-0001/ -- which I recommend you
read):

"""
We intend PEPs to be the primary mechanisms for proposing new
features, for collecting community input on an issue, and for
documenting the design decisions that have gone into Python.
"""

This is clearly a case of documenting a decision.

For future reference, questions about PEP process is what the PEP
editors are for.

> It also might be actually useful if the comparison page can be easily
> updated by a broader community, so that it will stay a valuable guide
> to all the projects looking to switch sometime soon.

Perhaps, but that's secondary to our purposes (and can happen in
parallel). Having a reliable & accessible historical document is
important here. The wiki page can be changed over time, and would no
longer record the thinking and data behind Python's DVCS decision. One
would have to delve into the page history, which is a pain -- and
won't show up in web searches.

-- 
David Goodger <http://python.net/~goodger>