On Thu, Jul 25, 2013 at 07:25:32PM +0200, Lennart Poettering wrote: > > Right now, we have a unified system which we pretty much guarantee cannot be > > targeted at all. It's moving too fast at every level. > Honestly this is the only thing that holds together Fedora at all. The 6 > month release cycle and the fact that the entire distro needs to be in > shape then is the only thing that keeps Fedora from falling completely > into pieces. I agree that that cohesive cycle is a strength -- there's a point somewhere in the slides where I mention that a shared release cadence should stay. > I am fine with splitting out the actual enduser apps out, but that's > nothing that can happen before we actually have a sane concept of > apps. But for the rest we should work on creating one strong unified > platform rather than a conglomerate of puzzle pieces that won't fit > together. You just weaken the name of Fedora that way, we won't stand > for anything anymore but a set of awkwardly non-integrated unsynced > components. Sure. A "conglomerate of puzzle pieces" isn't the proposal. That basically describes what we already have, although we're attempting to hammer and glue all of the bits into place. Another thing that isn't particularly clear from the slides is that the ring model represents divisions of policy, not necessarily OS parts (although, roughly the OS parts do divide in that way). I'll try to make that more apparent. -- Matthew Miller ☁☁☁ Fedora Cloud Architect ☁☁☁ <mattdm@xxxxxxxxxxxxxxxxx> -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
