From: Ben Finney on
Daniel Fetchinson <fetchinson(a)googlemail.com> writes:

> A good example for the first couple of stages of this process is PEP
> 3143 concerning adding a daemon package to the stdlib:
> http://www.python.org/dev/peps/pep-3143/

Thanks for the shout-out for PEP 3143. I can certainly say that the
process Daniel describes was a very healthy one for the development of
the API in that specification. Trimming the fat from the implementation,
and delegating concerns outside the specification, was a major benefit
of going through that public discussion process.

The PEP 3143 reference implementation has a lot of happy users, who are
making use of the API to perform the tasks they need. Once I complete
the extraction of dependencies, I'll be submitting the resulting code
for inclusion in the standard library.

> I haven't found the beginning of the thread discussing this

For the process you outline, the thread where I began soliciting
feedback of the “what's the best way to do this?” kind begins at
<URL:http://mail.python.org/pipermail/python-list/2008-September/560437.html>.
Some later threads gave me a better idea what was required and what was
possible.

While making a reference implementation, I drafted a PEP to describe the
interface I wanted the standard library to provide. Once I had it in
good shape, and had a reference implementation ready, I submitted it and
it was registered as PEP 3143.

The first PEP 3143 discussion thread starts at
<URL:http://mail.python.org/pipermail/python-ideas/2009-January/002529.html>.
Once I had incorporated a lot of the feedback, I fished again with
<URL:http://mail.python.org/pipermail/python-list/2009-March/1197730.html>.

It's annoying that many threads get broken because of poor transmission
of message references. You'll need to browse manually to get more
complete versions of some of the thread.