[SciPy-dev] Scipy workflow (and not tools).

Robert Kern robert.kern@gmail....
Tue Feb 24 18:45:43 CST 2009


On Tue, Feb 24, 2009 at 12:59, Matthew Brett <matthew.brett@gmail.com> wrote:
> Hi,
>
> I've split this off into a new thread because I felt there were two
> issues in Stefan's original thread.
>
> This is in the hope that we can stimulate discussion on the workflow
> (as opposed to - say - which version control system to use, or which
> bugtracker).
>
> I would be very interested to see if we can come to a consensus on the
> important discussion of whether to introduce fairly formal code review
> into the scipy workflow.  I've appended the key piece of discussion
> below.

My feeling about workflow is similar to my feeling about tools: the
experimentalist in me is back in the corner with his hand raised high
in the air. (He's very enthusiastic. "Ooh! Pickmepickmepickme!"
There's one in every class. You know the kind.)

There are a large number of unsupported assertions, gut feelings, and
common sense flying about, and I don't want to get any of it on me.
These are fairly poor guides for predicting the effects of project
policy decisions, especially common sense.

So let me make a meta-proposal: Let's do a series of one-month trials.
We'll pick a workflow to try for a month. Much of the opposition to
the various suggestions are coming from people who haven't tried to
work in the proposed environment (at least not with this group of
developers and this project; I conjecture without proof that variation
between groups and projects is are large factors in the differing
success of policies). The policy would be strictly enforced (to the
extent that it involves enforcement) for that month. Detractors from
the policy will grin and bear it for the duration of the month.
Because it's just a month. We'll try their idea next month.

This is far from scientific; the end result won't be measurable, per
se. But it will give us experience with each of the suggestions.
Perhaps what we fear about a policy won't be nearly as onerous as we
think, or even has the reverse effect. Maybe we'll generate better
ideas as we play around. Maybe some ideas truly suck. At the end, we
may not generate a true consensus, but I suspect that we'll all be
happier with *one* of the solutions than we are going to be if we just
talk about it. And our happiness is really the thing to optimize,
here, objective reality be damned.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth."
  -- Umberto Eco


More information about the Scipy-dev mailing list