On Fri, 12 Nov 2004 15:35:01 +0000, Douglas Furlong <douglas.furlong@xxxxxxxxxxx> wrote: > Jef was being incredibly unfair, there is actually some one on this > list, one of the Rodolfo's (sorry for the spelling?), who has been quite > vocal trying to get a minimal install up and running (remember the > argument about synaptec and up2date any one? :)) Several people have been vocal... and in the past 2+ years several people have gotten excited by the idea, and then wandered off never to return with something other people can test. Maybe I missed something, but I haven't seen an alternative comps.xml file show up in the devel or test list for community to test. Maybe the discussion has moved to the anaconda list. If you have a reference to something testable please give me the citation. A testable comps.xml that can be used to regenerate installable anaconda based images. We can spend days and days and days discussing what should be and what should not be in minimal install, trying to find the perfect package list...and all that discussion is absolutely pointless, if no one delivers a comps.xml file for other people to test. Waiting for someone inside Red Hat to make it a priority to rework the changes into a comps.xml is going to be a long long wait. Everybody who thinks this is important need to start delivering a comps.xml group file that can be tested. It doesn't have to be perfect, i expect several iterations of feedback about the finer points, but if someone is working on a new minimal grouping it needs to see a significant amount of community testing while its being developed. > Check the archives, there IS some one working on this, working hard I > believe, trying to work out what can be removed. Bah... show ME a comps.xml that I can test by rolling new installer images with. Talking about it only gets so far. And its been discussed several times now on the general lists. If someone in the community is hard at work on this, they need to start producing comps.xml revisions for community testing. -jef