[Beowulf] Oh.. IBM eats Red Hat
Robert G. Brown
rgb at phy.duke.edu
Mon Oct 29 09:47:31 PDT 2018
On Mon, 29 Oct 2018, John Hearns via Beowulf wrote:
> Bob Brown! Good to see you on here again Sir!
> > or get overtaken by cheeky
> > youngsters who pointlessly rename something (yum -> dnf, anyone?) and
> ... systemd... ?? arrggghhh? ARRGHHHH! It is Haloween on Wednesday... guess I
> should dress up as systemd, but that is far too scary,,,,
I don't know what to make of systemd as a design decision. I'm an Old
Guy, so by definition I grew up with init and the classic Unix OS
structure -- I still have all of the books in my office, sadly at least
semi-obsolete within the current kernels and linux layout. I also have
found it to be a moderate PITA to manage daemons and startup processes
using systemd because so much stuff is now even further under the hood
and more difficult to fix by just opening a file and editing it -- at
the very least, there are a lot more files and more protection.
But hey, I can cope, and have coped. I don't feel like I know the OS
anywhere nearly as well as I knew it back when I hacked around in the
kernel and wrote daemons myself, but it still "works".
The dnf thing rankles because yum was developed at Duke by my good
friend, the late Seth Vidal. He was hit while riding his bike through
Durham and died while he was working for RH as (I believe) the head of
the rpm/yum/installation team. If he had lived I don't think anybody
would have changed the NAME of yum to something as totally pointless (it
means even less than yum did -- at least yum paid homage to its open
source roots) as "dnf" while essentially leaving its purpose and
functionality unchanged and while leaving all of its internal plug ins,
initialization files, and so on named with "yum". Pure ego bullshit,
but then, so was gnome3.
The point is that open source maintenance is still a thorn in the side
of the process. In order for a tool to survive, somebody has to own it
and contribute AT LEAST enough time to keep it compiling and
functioning, forever. And there are tens of thousands of tools -- I
haven't even attempted a count of the total number of packages in debian
or fedora in years now but there were close to 20K in fedora last time I
did. That's a lot of people who have to be maintaining things for the
distro(s) to remain coherent, and almost all of those people have day
jobs or just get old and life intervenes. Or ends.
Hence the importance of having somebody with real money -- RH, IBM, I
don't care -- maintaining a core of developers and providing a stream of
support moneys to keep the whole thing coherent and functional over the
long haul, WHILE the kernel people do things like insert systemd, or the
font people completely revamp the fonts (breaking tools like
xfig/transfig in the process that then have to be fixed even though the
original developers are long gone and the libraries they use are X
libraries that are THEMSELVES on the chopping block...
> > then yank it around) and somehow manages to block CENTOS -- mixing in
> > more proprietary stuff, for example so that CENTOS is basically cut off
> > from the development stream of key new packages -- then the free
> >software world we live in will get nastier and less stable.
> SOrry, no CentOS will nto be throat choked, or will nto be if IBM sees any
I certainly hope you are right, and am optimistic as well. But your
reply does contain the critical conditional: "if IBM sees any sense".
In the end, it will come down to money -- will IBM see sense in
continuing an investment in something that does not directly contribute
to the bottom line and that might even subtract from it by providing a
free alternative to something they sell? Will IBM recognize the
importance of supporting "the community" while it sells the fruits of
that community for enormous markups (it starts out "free", remember) to
shirts in banks? Does IBM "get it", that support for the broad OSS
community IS investing in its critical development chain, that it isn't
about closed room proprietary development and software copyrights and
patents and so on?
All we can really do is hope so, and/or hope that we continue to have a
few alternatives to JUST debian in the kitchen sink, free as in beer
> Remember Karanbir Singh has a position with Redhat (maybe a board position).
> I hope he got lots of stock options - and if so he owes me a beer or two
> next time I see him at FOSDEM.
> Remember what CentOS is - it is a clone of Redhat, which is used by those
> folks who will not pring for RHEL licenses. They prefer to do their own or
> community support.
> CentOS is there to keep mindshare for RedHat (and now? by extension IBM).
> Cut off CentOS and all those academics and supercomputer guys drift towards
> And remember Ubuntu -Canonical dis not become popular amongst the hip cloud
> kids and the biotechnology types because of word of mouth from someone with
> a big beard and artisan coffee. Ubuntu was carefully marketed and propoted
> to appeal to the cloud community - and we now see the results.
> I vouch that many cool kids dont even know there IS another Linux.
> SO back to CentOS - it costs virutlaly nothing to Redhat to make it
> available - and they types who use it aint gonna suddenly start springing
> for RHEL licenses, so they lose nothing but gain loads by having it
> AS an aside, I Was discussing RHEL on two big clusters with my friends
> regarding IBM. Redshat used to have an HPC license which was equivalent to
> Redhat desktop. SO if you had thousands of nodes ti was nto TOO expensive.
> IS that HPC License still in operation?
> On Mon, 29 Oct 2018 at 15:31, Prentice Bisbal via Beowulf
> <beowulf at beowulf.org> wrote:
> You mean this?
> In 10/29/2018 10:56 AM, Peter St. John wrote:
> I know that several years ago, more than a decade, IBM was
> selling a solution of running some thousand instances of
> Red Hat on a 390. I don't know how that competed with
> racks of commodity etc but I can imagine there were
> advantages in the backbone. Anyway they were selling it
> way back, which at the time I thought was progressive. But
> I'm just an application developer.
> On Mon, Oct 29, 2018 at 10:52 AM INKozin via Beowulf
> <beowulf at beowulf.org> wrote:
> exactly my thoughts (even though i have not worked
> there, talking to its employees was enough). it's
> attitude towards open source is not exactly
> the recent github deal comes to mind but at least MS is
> declaring to be more open towards open source.
> and at least there is an alternative in that case -
> what would be an alternative to RH? certainly not a single
> On Mon, 29 Oct 2018 at 07:43, Tony Brian Albers
> <tba at kb.dk> wrote:
> I wonder where that places us in the not too
> distant future..
> I've worked for Big Blue, and I'm not sure the
> company cultures are
> compatible to say the least.
> Tony Albers
> Systems Architect
> Systems Director, National Cultural Heritage
> Royal Danish Library, Victor Albecks Vej 1,
> 8000 Aarhus C, Denmark.
> Tel: +45 2566 2383 / +45 8946 2316
> Beowulf mailing list, Beowulf at beowulf.org
> sponsored by Penguin Computing
> To change your subscription (digest mode or
> unsubscribe) visit
> Beowulf mailing list, Beowulf at beowulf.org sponsored by
> Penguin Computing
> To change your subscription (digest mode or unsubscribe)
> visit http://www.beowulf.org/mailman/listinfo/beowulf
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
> To change your subscription (digest mode or unsubscribe) visit http://www.be
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin
> To change your subscription (digest mode or unsubscribe) visit
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
More information about the Beowulf