On Thu, Aug 16, 2001 at 10:15:50PM +0100, Nicholas Clark wrote:
> We've missed something fundamental - there is no safety whatsoever in
> re-rsyncing during a testing run. The rsync server may have moved on.
> If you rsync to a later patch midway through a series of tests you ruin
> the experiment, as the whole idea of the thing is to build the same
> source in a variety of configurations.
Second problem, we're all smoking against different patch levels. To
make this really effective, we should all work on the same one. I
know there's some sort of daily snapshot lying around. Insted of
rsyncing perl-current, we should probably download this. It'll also
lose the dependency on rsync, which doesn't work on alot of OS's.
--
Michael G. Schwern <[EMAIL PROTECTED]> http://www.pobox.com/~schwern/
Perl6 Quality Assurance <[EMAIL PROTECTED]> Kwalitee Is Job One
BOFH excuse #55:
Plumber mistook routing panel for decorative wall fixture
