On Thu, Jul 27, 2000 at 06:23:16PM +0300, Jarkko Hietaniemi wrote:
> You may want to think about psychology and wording, though. If
> someone writes a code patch but does not supply tests/docs, and he or
> she is told that the patch is rejected because of that deficiency, he
> or she might not submit patches ever again. Writing good test cases
> is much harder than the actual fixing itself, but we still should
> give positive feedback to the fixer for his or her efforts.
Yes yes yes! Instant rejection of patch submissions bad. Instead of
rejection, we may want to establish a holding area for incomplete
patches. It will have to be very careful wording to make sure they
understand the patch is waiting for completion as opposed to
rejection. I'm probably too blunt to write those messages.
We'll also need to set up a set of test patch writers/mentors to help
new code patchers along in authoring tests. There is the problem of
teaching how to write a proper test, since I think there's little
practical experience out there, and also convincing people that the
tests are necessary and we're not just being anal-retentive.
--
Michael G Schwern http://www.pobox.com/~schwern/ [EMAIL PROTECTED]
Just Another Stupid Consultant Perl6 Kwalitee Ashuranse
Plus I remember being impressed with Ada because you could write an
infinite loop without a faked up condition. The idea being that in Ada
the typical infinite loop would be normally be terminated by detonation.
-- Larry Wall in <[EMAIL PROTECTED]>
