[CentOS] Vote For CentOS :)

Sat Jun 4 02:19:03 UTC 2005
Lamar Owen <lowen at pari.edu>

On Friday 03 June 2005 08:48, Johnny Hughes wrote:
> I am a little touchy on this subject ... being that I hear it so often.

I most assuredly understand this.  I reply to your message in this thread 
because I find you quite level-headed.

> I disagree ... the work that RedHat put into RHEL is significant.  It is
> a great distro, and worthy to win this award and many others ... but so
> is CentOS.

> BUT ... they didn't put any work into CentOS.  All their work was put
> into RHEL.  They are compensated fairly well for that work too, I might
> add.

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 
of the changelog entries in the various spec files in CentOS what percentage 
of persons making changelog entries have an @redhat.com address?  I have some 
work in CentOS in the PostgreSQL set, but it is quite small compared to the 
rest of the work.  But Red Hat developers (some of who are upstream 
developers paid to work on the upstream package by Red Hat) have put a 
substantial amount of nontrivial work into the packages that through the 
CentOS team's efforts became CentOS.  

I do understand what you are saying, why you are saying it, and the 
distinction you are attempting to make, but it comes across at least to me as 
understating the work Red Hat people have done.  Let's not overstate Red 
Hat's contribution, by all means; but let's ot understate it either.

> the items they publish from somewhere else. They are required,
> therefore, to make their source code public.  They take that requirement
> seriously, and they do an outstanding job of publishing their source
> code openly.  They should be commended for that.  I do it every chance I
> get :)

They are not required to make the source for non-GPL code public (unless the 
license requires it; the PostgreSQL BSD license for instance does not require 
it), nor are they required to distribute any spec files or special 
initscripts they may use.  They are not required to distribute source to 
anyone but the receipients of the binary code, and they cannot restrict said 
recipient's right to redistribute the source under the GPL.  But GPL code is 
not the majority of the Red Hat dist, is it?  Anyone have a count of bytes 
under GPL versus other licenses in CentOS?  If not, I'll do the math, 
tomorrow.  So Red Hat could make the job much more difficult by distributing 
the source as simple tarballs with no specs, no initscripts, no notes.  Like 
you say, they have done an outstanding job releasing things they don't have 
to release as source in SRPM format, making it not as difficult as it could 
be to rebuild a trademark-free build.  Not that it is trivial or even easy; 
but it could be more difficult.

> Using your logic, The only people who would get credit are the
> programmers who write the code for the parent projects.

Many of whom work for Red Hat.  Some of whom could not afford to work fulltime 
on their project were it not for Red Hat's employment.

> Or the Fedora 
> Core volunteers who actually package and test probably 95% of the stuff
> that get into RHEL.

Most of the Fedora Core packagers work for Red Hat, with notable exceptions.

> So no, I don't think that you should vote for RHEL in a place where
> CentOS is also listed 

I agree wholeheartedly with your conclusion, but not with all your reasons.
-- 
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC  28772
(828)862-5554
www.pari.edu