Hi, On Sonntag, 13. Februar 2011, Bernhard R. Link wrote: > I'd rather say "doing it right" means to properly test it's build to be > robust. Only ever testing in an artifical environment has a certain > outcome: certain failure.
Nobody said people should stop rebuilding packages on those 100 derivatives we
already have or stop creating random backports in random environments (or or
or). And nobody said we shouldn't do other systematic rebuilds in known
broken environments. So, this kind of testing will not stop.
It's still completly beyond me, why you think it's useful to randomly build
1/12 of our binary packages in a random environment, ship that as stable to
the world and not even make clear which those packages are, plus throwing
away the build information too. It's a mess, and a harmful one. As you
observed, it might have side benefits sometimes, but those can be achieved by
other means, which don't have the downsides discussed here.
EOT for me.
cheers,
Holger
signature.asc
Description: This is a digitally signed message part.
