On Sat, Aug 15, 2009 at 09:14:37AM -0400, Michael K. Johnson wrote: > I'd like to make a few points clearer based on off-list discussions. > I'm not saying that the work that António is doing on an import > fully built from source has no value. I'm merely saying that it is > not clear to me that it has value over a binary import, and I want > to make it clear that there is an option here that it is not clear > to me has been considered, because it's not clear to me that the > Foresight development community is even aware that it is an option.
i don't think a binary import option has been considered, at least i would not consider it. while it is not clear to you what the value of a from source build is over a binary import, the oposite is also still very unclear, especially if we are considering a mixed import. any binary import locks us in to that version, if we want to update any individual package that affects some binary import, then we have to throw away the affected binary and rebuild it from source. the downside of that is that at that point we need to do more work, whereas had that affeced package been built from source from the start, then it would be a mere rebuild away. so i really do not see the advantage of a binary import, we still need to test our results in -qa just as well. if we can not rebuild arbitrary packges because we need to update a core library in order to eg build the latest gnome, then this will create a problem. also since conary always tries its best to satisfy dependencies we'll just end up hunting for which binary it is that prevents us from updating some library. to me this just feels very messy... > If the from-source rebuild is the easiest to maintain and provides > clear value that furthers Foresight's core differentiation, great. i see a core differentiation in the fact that foresight will be an upstream distro that others build from. along with debian, gentoo, fedora, and others. if we make fedora our upstream then people will expect a fedora like system and not see foresight as an alternative. i believe our situation is manageable. what needs to happen is that ken and antonio need to delegate more tasks, even if that means that it will take longer for that task to be done. others simply need to learn how to take care of these things, and ken and antonio need to be on top managing us. we are your workforce. use us! greetings, martin. -- cooperative communication with sTeam - caudium, pike, roxen and unix offering: programming, training and administration - anywhere in the world -- pike programmer working in china community.gotpike.org foresight developer foresightlinux.org open-steam.org unix sysadmin iaeste.(tuwien.ac|or).at caudium.org Martin Bähr http://www.iaeste.or.at/~mbaehr/ is.schon.org _______________________________________________ Foresight-devel mailing list Foresight-devel@lists.rpath.org http://lists.rpath.org/mailman/listinfo/foresight-devel
