Editor's Note: That Which Does Not Kill Us...
Oct 27, 2006, 22:30 (14 Talkback[s])
(Other stories by Brian Proffitt)
Re-Imagining Linux Platforms to Meet the Needs of Cloud Service Providers
By Brian Proffitt
I was once offered to write a small book about Nietzsche. I
declined, opting instead to accept another writing assignment about
Plato. My reasons were purely personal; Nietzsche depresses me. I
don't refute his works, per se. I just tend to avoid them.
But there is one quotation that is attributed to him that I think
is perfect for this week's news: "That which does not kill us makes
I'm sure that phrase is running around Red Hat headquarters
today. There is some truth to this aphorism, and it needs to be
remembered in the days ahead. I think we are in for interesting
times ahead. Like the Chinese curse "may you live in interesting
times" kind of interesting.
I know a number of people, including some of my colleagues in
the open source community, were really angered by the audacity of
Oracle when CEO Larry Ellison announced that his company would
"fully support" Linux. His definition of support: take the purely
open source code of Red Hat Enterprise Linux, strip anything
copyrighted from it (like nomenclature, icons, etc.), and release
it as part of Oracle's stack. RHEL becomes Unbreakable Linux.
I have to admit, my first reaction was a bit sardonic: it's no
secret that Red Hat has ticked off many partners and customers with
what can only be described as their arrogance. Red Hat was the
number one commercial Linux offering, and by golly, everyone was
going to know it. Every time I turn around at a Linux event,
someone's bending my ear with a "look what those $@#%&! have
done now" story. And, I think, I was not alone in this initial
assessment. Red Hat has definitely alienated quite a few people in
the open source community, and it was interesting to note that the
average reaction from Joe Community to Red Hat's plight was
Thinking about it further, I realized that ultimately this was a
really great thing for Linux and the open source commercial
How many times have you been asked by some business school grad:
"okay, if I give away the code, how do I make money on it?" The
answer is always a variation of this theme: "With superior services
and support. Now go away, kid, you bother me."
And here we are, on the sidelines of history, because never has
there been such a clear and unadulterated example of this theory
put into practice. Oracle is undercutting Red Hat's support pricing
model, but will they offer the value to their customers that Red
Hat can? And can Red Hat get that message out? Other pundits,
perhaps wiser than I, are addressing this concern and I won't
re-hash their words here. Suffice to say, this is it. This is where
the rubber meets the road and we can all find out, on a level
playing field, how viable a commercial model open source really
What is even more definite, and exciting, is the fact that, no
matter what happens to Red Hat, Inc., there is no disputing that
Linux as a business platform has well and truly arrived. Granted,
we've been saying that for years. But even the most jaded Microsoft
customer is going to have to notice Oracle's in-house support of
Linux and wonder.
Go on, I dare someone to call Linux the "upstart" operating
system again with a straight face. I will be the first to
So, the open source commercial model gets its first real test
and Linux is validated for all the world to see as a real
technological choice. Is there a downside to all of this?
I was pretty optimistic about all of this until I woke up this
morning and started picking up the thread out in the media about
Red Hat being a prime acquisition target now that it's stock values
have been hammered down so far. Then the little bell went off in my
head. Maybe the commercial open source model has a flaw, after
Here's what's troubling me. It is a given that open source's
model can be a commercial success. But there is a disturbing trend
in the commercial vendor world that I am seeing. The open source
model doesn't seem to be able to reach any sort of conclusion. A
successful model should show an open source company start small,
build quickly, then reach a stable rate of slow growth as the
company continues on into comfortable middle age. To grow more
quickly, it makes sense to partner with more established vendors.
If I am a bread vendor, then getting a partnership deal with
McDonald's is an excellent way to hitting the big time.
The problem I see is that after a certain point, whenever a
small-fish open source company gets successful, it seems that they
get snatched up by a larger fish. Ximian, SUSE, and JBoss are three
recent examples of this. Is Red Hat going to be the next
I'm sure some will point out that this happens all the time in
the business world, regardless of whether a company is using open
source or not. I know that. But the potential for Oracle (or some
other company) picking up Red Hat may be the first time we have
seen open source being used as a deliberate weapon against the open
source company. That goes beyond my initial assessment of two
competitors on a level playing field scenario. Now the very premise
of open source can potentially be used to cut a competitor down to
their knees, weakening them to the point where they are more easily
dispatched... or acquired.
After all, Oracle acquiring Red Hat now makes some sense. Many
are already questioning Oracle's ability to deliver decent support
for their new Linux offering. What better way to pick up the very
best Linux support by buying the same company you copied the open
source code from? A company which is now priced at a bargain level,
thanks to jumpy stock holders.
It's not just Oracle. A friend of mine suggested to me this
afternoon that IBM, ever ready to expand its services business,
might find Red Hat too tempting a target. That argument makes a lot
of sense, but time will tell if that plays out.
Has Oracle found the chink in open source's commercial armor?
Maybe so, but one thing to keep in mind that if Red Hat does
survive because it adjusts to this new competition (and I believe
it can), then this trick can't be tried more than once. Nor can
just anybody do it. Look at Oracle: they've been partnering with a
Linux company for years and already their claims of being able to
handle Linux support are being met with a large degree of
What would happen if Microsoft, say, announced it was going to
repackage Red Hat code and offer support for it? They'd be a
laughingstock, and with good reason. So this kind of attack really
only works if the company offering the relabeled product/support is
a credible Linux source. This is not Kryptonite.
Bear in mind, that I am only referring to open source offerings
in a commercial sense. If Red Hat the company dies tomorrow, Red
Hat Linux would live on in some way. As would any other product.
The commercial viability of an open source company is what I am
pondering about. Is the ability to legally copy code and repackage
it going to be a useful weapon against any one open source company
getting too big? We're about to find out.
Nietzsche's words are really ringing true. Whether you like Red
Hat or not, you should take great notice of how they come out of
this. It could be the shape of things to come for commercial open