Hello all--
I recently successfully rebuilt RedHat 3.0AS, and have a distro that is installable by ftp/http/nfs etc...
We have the cAos work, the White Box build, and certainly other work all towards a RH enterprise "rebuild". From the discussions it seems quite clear that this "rebuilding" business is nontrivial, with a fair amount of work going into making the packages actually build - on some system.
What I wanted to ask some of those involved is how much more work it might seem to them to be able to build and distribute a system that would ACTUALLY BUILD with no problems, gotchas, etc. ON ITSELF?
What I'm thinking is that the problems with such simple rebuild efforts seem to me in some sense to put a lie to the "open source" notion. That is if the idea (an idea?) of "open source" is that you get the sources and, if you need to, you can make minor modifications to the source and rebuild it to suit your needs (as per Stallman's initial Xerox motivation), then if the source build doesn't work as distributed, where is that value?
This seems to me an area where the RHel rebuild effort might, at relatively little cost (?), make a significant contribution to the open source movement. I.e. this Rhel rebuild system could be more truly open source than other distributions - e.g. Redhat and others - and (?) still include the added value of tracking Redhat and conveying the value of the Redhat integration efforts and a home for the Redhat refugees (e.g. like me).
I'm thinking that essentially marketing a rebuild with this added value (builds on itself from source RPMs!) would add appeal and draw attention to such a rebuild distribution. That added value might make it more of a gathering point (I'm looking for a new Linux "sweet spot" to replace Redhat who seems to have abandoned the position to make a profit) for Linux distributions.
One concern I have is that such an effort might in some ways compromise the Redhat compatibility. If changes had to be made to make a particular package build on the system itself, then might it be that future errata distributed by Redhat would no longer work as updates? Still, it might also serve as a justification for a fork in the road point to a new sweet spot distribution (e.g. as per some of the initial cAos goals).
Thoughts?
--Jed http://www.nersc.gov/~jed/
rhel-rebuild mailing list [EMAIL PROTECTED] Hosted at the University of Innsbruck, Austria
