[Zope3-dev] Development methodology (Re: [Zope-CMF] Future CMF) (rant)

[R. David Murray]

On Mon, 7 Oct 2002, Lalo Martins wrote:
> Have you noticed how Zope3 development has slowed down a bit?
>
> IMO the community development of Zope3 (and other Zope-related projects in
> general) is all wrong. The metodology is too documentation-oriented, there
> are lots of virtual-paperwork involved in projects that otherwise would be
> simple.

I don't think the documentation requirements have anything to do
with it, personally.  It certainly wasn't why there was a big gap
in doing anything for me, and I doubt it influenced SteveA or Gary
Poster any, either (just to speak of two who contributed an bunch
but have or had paused in their contributions for the moment).  It
seems to me as though those of us actually working on Zope3, the
current process is working pretty well.  I don't feel like I have
to submit any virtual paperwork other than the stuff I *want* to
submit in order to feel like what I am doing is going to play will
with Zope3 in the opinions of the other developers.

The use cases and Interface specs are as much a way of *thinking*
about the problem (publicly) as they are a documentation requirement.

> Community-oriented open-source development is not compatible with ISO9001.

True.

> And of course Zope3 development can't even tap the "need to fulfill" pool at
> all, since it is hardly usable to fulfill any real-life need.

True, but the more documentation we have when we *do* get to that
point the faster those additional contributions will come in.  And
despite the doc requirements you bemoan, Zope3 is *not* documented
well enough yet (IMO).

> Additionally, most developers have little to no interest in commenting other
> people's ideas before there is some code to test.

I don't think this is true, either.  But I do think that most people
aren't interested in commenting unless it impacts on something they
care about.  I don't do TTW content development anymore, so the
example Proposal you mention is not something I had any motivation
to comment on (for example).  I suspect the dearth of comments has
more to do with the same factors that have slowed down Zope3
development rather than being a cause of the same.  Either that,
or Jim got it right the first time <grin>.

> duplicates effort but isn't at cvs.zope.org) than to make proposals on CMF
> (or Zope or Zope3 or whatever) is a symptom that there is too much
> bureaucracy involved. And I say this with the authority of someone who just

For Zope3, I don't think this is true.  For Zope2, I think the
bureaucracy/process is still a work in progress, and *does* need work.

> My suggestion is that we move to a model more similar to the PEP system. The
> *first* artifact necessary for a project is a prototype. If the author
> doesn't yet know the details, it's ok to raise a discussion on the list or
> IRC, but then it isn't yet officially a "project", just a discussion.

If I *had* to produce a prototype first, I'd still produce some sort
of "plan" document first that I'd post somewhere to get comments before
I put too much effort into a prototype.  Unless the prototype was
simple to code, of course.

If you prefer, you can always code a prototype first, and release
it and the documentation about it at the same time.  So I don't
really see the *functional* difference.

Speaking from my own recent zope3 experience, I coded a proof-of-concept
prototype first, then posted a proposal, and am now working on the
Real Thing.

--RDM