[Beowulf] Oh.. IBM eats Red Hat

[Robert G. Brown]

On Tue, 6 Nov 2018, Gerald Henriksen wrote:

>> I though Fedora was killed of a couple years ago after version 7.
>
> Red Hat has never (publicly at least) pondered killing Fedora and
> still continues to invest significantly into it.
>
> Of the 2 Fedora is likely more vulnerable because its utility to a
> server version of RHEL isn't as clear and it is unlikely that it would
> survive without the funding (direct and indirect) of Red Hat.

Yeah, but Fedora ``is'' RH's development process.  RHEL is deliberately
hyperconservative and can't indulge the inevitable instability of the
rapid development and turnover in the OSS community.  On the other hand,
it can't afford to freeze forever and just stand still.  Fedora is their
alpha, beta, and deployment development stream.  By the time something
has been in Fedora for several releases, it will usually be stable
enough to be considered for inclusion in RHEL's next release, and RH has
a very good idea of its popularity and usage rate from just looking at
the fedora yum/dnf logs.

Fedora is indeed highly vulnerable from the point of view of funding,
but Centos is much more of a direct competitor because the rapid pace of
releases and the limited support life for fedora makes it less than
ideal in the server room and out of the question for major corporation
server rooms that require super-stability and extreme security (banks,
for example, have to have their server room software audited half to
death to get approval to use it at all, which costs money so they would
be happy to have a functional package approved and have it last
"forever" without change).  Centos IS RHEL, and only the honor system
keeps it out of midrange server rooms and off of development desktops
maintaining RHEL server rooms.

It will be very interesting to watch this unfold.  As I've said, I'm
pretty optimistic, because in my own experience IBM has been
linux-friendly and M$ hostile for twenty years, and I really don't think
they are out of touch with the OSS culture and development cycle.  It
may come down to whether the geeks or the bean counters win in some sort
of fight over money flow, but I honestly think that the IBM board is
likely to be smarter than to let that happen.  (I think that) they
bought RH as an investment in the long game, not to strip it of its
assets and cash flow, rebrand it, and dump both it and all of its
foundation and development cycle.  They have to recognize that even if
they DID dump fedora, they'd just have to implement their own in-house
alpha/beta QR cycle, but now feeding off of a non-RPM stream from e.g.
Debian.  They'd end up duplicating all of the core infrastructure that
Fedora has now except the distribution side, and would piss off the OSS
community enormously in the process.

They have quite literally nothing to gain by pissing off the community.
The reason RH IS a major player in the server room is because a large
fraction of computing professionals adopted linux for their own personal
use, are fully aware of its capabilities and skilled at its management,
and advocate for it successfully against M$'s massive marketing team and
its FUD.  The last thing in the world IBM needs is to convince those
people that it is becoming a M$ clone; quite the contrary.  If anything,
they should work hard to make those people feel even BETTER about RH/IBM
together than they already felt about RH alone.

    rgb

> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
> To change your subscription (digest mode or unsubscribe) visit https://urldefense.proofpoint.com/v2/url?u=http-3A__www.beowulf.org_mailman_listinfo_beowulf&d=DwIGaQ&c=imBPVzF25OnBgGmVOlcsiEgHoG1i6YHLR0Sj_gZ4adc&r=VjN6W5gT-iXGupU3t6I7tA&m=QCmzOHy4vxRR4Gz1YAYwc_7dyCKNfzSlb7UANmxizzo&s=TUJLH26RhJAY8p4r4q5XJvWaGOT7a7G-QB3SFuIT380&e=

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