Hi Troy,
* Troy wrote 13 Aug 2005:
> Fabian Braennstroem wrote:
> >Hi,
> >
> >I just saw scientificlinux on the net and wonder where the
> >advantages compared to other distributions (e.g. centOS)
> >are. I suppose that I can use the same packages!?
> >Is it more suitable for scientific work like the name
> >implies?
> >
>
> This is a question I think we'll need to put up on the FAQ page. It's a
> good one, that get's asked enough. But for now, here is at least what I
> feel the answer is.
>
> First off, we believe that CentOS and Scientific Linux are both
> compatible, for the most part. We don't guarantee that, but since both
> distributions strive to be RHEL compatible, and we believe both have
> achieved that goal, we feel that they are compatible.
> Second off, we do not feel they are 'competitors'. (I know that wasn't
> in your question, but others ask it) Many of the main developers are on
> the same mailing lists, and we try to help each other out whenever
> feasible. There is cooperation between our two distributions.
>
> So, to the main question, what is different.
>
> One of S.L. goals is for a person to install the release they want
> (let's say S.L. 3.0.2) and be able to sit at that release and know that
> nothing is going to change except security errata. That is why you will
> see that we still have 3.0.1, 3.0.2, etc... This is because several of
> the experiments or labs have tested 'this particular release' whatever
> that release is, and they don't want anything changing on it.
> CentOS is different in that their releases get upgraded to the 'head'
> release after it is released. So if you installed what we consider
> 3.0.4, and then when 3.0.5 came out, you would automatically be upgraded.
> We aren't saying one is better than the other, just that that is a
> difference. We are putting things in place so that if a person wants to
> have S.L. automatically update to the latest they can, but that won't be
> the default.
>
> Another difference is the concept of 'sites' in Scientific Linux. This
> is something that is different ... from any other linux release that I
> know. It is a way for a 'site' (like a lab or a University) to put in
> various changes, like scripts, rpm's, installer changes, and customize
> their site, yet still retain S.L. compatiblilty and continue to get the
> security updates. This is how we create Scientific Linux Fermi, and I
> believe there are a few others that use this functionality.
>
> Our sponsors are different.
> Scientific Linux is mainly sponsored by scientific lab's and
> Universities around the world. Mostly by Fermilab and CERN. There are
> other's that contribute and help, most definatly. But those are our
> main sponsors.
> CentOS is a purely community based OS. All of their sponsors, their
> servers, bandwidth, and contibutions have come completely from the
> community.
> This difference tends to give each of us a different point of view when
> looking at various ways of developing our releases.
>
> Is Scientific Linux better for scientific applications?
> Basically it is called Scientific Linux because it is made by scientific
> labs, for scientific labs and universities. It is not named Scientific
> Linux because it has the largest collection of scientific programs. It
> was named back when it was small, and only the scientific labs were
> using it.
>
>
> >And a small OT question. If there are certain packages
> >missing, is it possible to build them using an existing
> >rpmsrc-Package from fedora and install it? I am actually
> >using archlinux right now and like the ability to build own
> >packages.
> >
>
> Certainly, although I would suggest looking to see if someone else has
> already compiled them. Check out our repository site
>
> https://www.scientificlinux.org/community/repo/
Very informativ, thanks! I will try it the next weeks and
province the admin at my institute :-)
Best Greetings!
Fabian
--
|