On Saturday 04 June 2005 01:27, Greg Knaddison wrote: > On 6/3/05, Lamar Owen <lowen at pari.edu> wrote: > > Um, I don't if I can agree with that statement. CentOS is built upon the > > source RPMs of RHEL with sometimes substantial modifications, right? So, > > out > Lamar Owen pays attention to what's going on here. Well, thanks. > If he thinks that there are substantial modifications - what is the > casual user to think? > Regardless of what is in the rpm name, the spec file, the mantis, etc. > their should be a better communication of what has or has not been > changed. The only way to really tell is to check the changelogs for all packages and diff that with a RHEL system. Assuming the changelogs are modified (which AFAICT they are). rpm -qip --changelog packagename.rpm. Iterate over 1,494 packages. :-) I used to provide lots of detail in my changelog entries (see the PostgreSQL spec files for the previous versions of PostgreSQL; the changelog historically has been truncated at each major version to keep the size down) so that people could fathom what I had changed from release to release. Tracking these sorts of changes either needs to be fully automated (Johnny is so right that it takes time to do this) or someone needs to step up to the plate to chug through the changelogs, spec files, and patchsets. A specfile changelog tracker might be nice; if the specfiles are in CVS (or other revision control system) a committers list could be set up that would e-mail that list on each CVS commit (lots of projects, including PostgreSQL, do this). Then it's up to someone to digest this for public consumption. I get these sorts of e-mails from the PG rpmfoundry CVS server (while not terribly active there anymore, I still track it and provide advice when asked). The practice of segregating the repo into a CentOS base, the addons, and centosplus is great; unfortunately there are changes that are in the base that really must be in the base. Such is life. As far as the 'substantial' part of things, the centosplus repo is what I'm thinking about, along with the building into the ISO form, setting up the install, etc. -- Lamar Owen Director of Information Technology Pisgah Astronomical Research Institute 1 PARI Drive Rosman, NC 28772 (828)862-5554 www.pari.edu