On Tue, 2008-11-11 at 12:51 -0600, Rex Dieter wrote: > Ralf Corsepius wrote: > > > On Tue, 2008-11-11 at 12:34 -0600, Rex Dieter wrote: > >> Ralf Corsepius wrote: > > >> <sarcasm> > >> Who really needs all that fedora infrastructure, peer/package review, > >> qa/testing, anyway. Much easier to skip all that. > >> </sarcasm> > > > Well, ... if the fedora infrastructure was really serving contributors, > > if peer/package reviews were functional, if testing was functional, then > > this all would not be an issue. > > > The unpleasant truth is: It isn't. > > Standard response: how exactly isn't it? Let me pick just 2 examples: * infrastructure: Fedora's infrastructure in first place mean struggling with a zoo of more or less arguable processes/work-flows (freezes, FTBP, wikis ...), a zoo of more or less functional tools (bugzilla, FAS, packagedb, koji, bodhi, ...) and a zoo of bureaucracy they are implementing. * testing: The parties testing a contributed package in first place is the package's upstream, the packager and this package's end-users. Fedora only contributes to testing a package insofar, as having a package in Fedora widens the "potential user-base" of a package. > How exactly do you propose to make it better? I guess you should know my answers :-) Points to getting started with would be * "getting rid of the freezes" * "getting rid of the update delays" * improve the tools contributors are forced to use. Ralf -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
