On Tue, 2007-06-19 at 15:29 -0500, Clark Williams wrote: > Jeremy Katz wrote: > > On Tue, 2007-06-19 at 00:40 +0200, Axel Thimm wrote: > >> On Mon, Jun 18, 2007 at 06:11:47PM -0400, Jeremy Katz wrote: > >>> The question is where in the process does it make sense to generate > >>> the deltas. Do we do them in the build system (the koji level)? Do > >>> we do them in the update system?[1] Do we do them at tree compose > >>> (mash) time? Each has tradeoffs... not sure which is really the > >>> best. > >> The problme domain are the limited bandwidth of end users, so you want > >> it to run on what end users consume. This can be different for > >> released versions vs rawhide, but in general released versions would > >> only perform this on updates-released (not updates-testing) and > >> for rawhide on everything. > > > > Yeah, it's easy to see that deltas for -updates and -updates-testing[1] > > from the release and probably the last update make sense[2]. rawhide is > > definitely the trickier to figure out where to draw the line. > > Do you even want to draw that line in rawhide? To me, rawhide is > dangerous enough that trying to add delta updates on top of it is just > asking for (more) trouble. > > I think that -updates* is perfect for deltas, because the latest package > is almost always going to be an iteration on the previous version(s). > But rawhide might be a complete version jump, or changing toolchains, or > trying something just plain whacky. The potential for space saving due > to commonality doesn't seem to be there for rawhide. The space savings benefit isn't really the win for rawhide -- it's more the testing base and being able to ensure that things continue to work. Because nothing sucks more than doing a release and then realizing that some part doesn't work because it's not a regular part of your devel infrastructure. Jeremy -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
