[Bioclusters] how are the Redhat product changes affecting existing and future plans?

[Tim Cutts]

On 04-Nov-03, Matthew Laird wrote:
> On Tue, 4 Nov 2003, Tim Cutts wrote:
> 
> > My gut feeling is Debian for bog-standard compute nodes, because it has
> > no commercial agenda, and will not be shot in the head because of such
> > agendas.  Plus, of course, the fact that it's a damned fine distribution
> > anyway (OK, I'm a bit biassed, I used to be a Debian developer).
> 
> Now don't get me wrong, I love Debian and have used it at home for years, 
> but the largest drawback of Debian is the fact they can't keep up on 
> hardware support and new version of packages the same way a commercial 
> distro can.  We've been discussing this around here and have been 
> wondering if the new Fedora project might suffer this same fate.  In fact 
> this situation could cause both Debian and Fedora to fall further behind 
> on hardware support if volunteer developers are now split between the two 
> projects.  

Debian's hardware support has never been an issue for me; hardware
support is a kernel issue, and Debian's make-kpkg makes building custom
kernel packages an absolute breeze.  The issue exists only if you're not
willing to roll-your-own, and even then, I wouldn't say they're too far
behind; their kernel is 2.4.22 now, I think.

Their glibc version is actually ahead of most other distributions, which
is a cause of some problems (LSF doesn't like glibc 2.3 much, for
example)

And, not relevant to the current discussion, their XFree86 support is
good, especially for non-Intel architectures, where they basically do
the port for everyone else.

Your mileage may vary, of course.

Tim

-- 
Dr Tim Cutts
Informatics Systems Group
Wellcome Trust Sanger Institute, Hinxton, Cambridge, CB10 1SA, UK