Subject: Re: update server alternatives
From: Tom Lord <lord@emf.net>
Date: Tue, 29 Oct 2002 11:58:41 -0800 (PST)




       >     Tom> They are trying to turn an FSB into a de facto proprietary
       >     Tom> company -- to exploit the weaknesses of the community to
       >     Tom> their advantage rather than trying to fix those weaknesses.

       Why is that bad for RH?  RH is a business.  The community has a
       weakness.  RH exploits that weakness to create a commercial
       opportunity.

As you might have noticed, I like to try to express complex ideas via
simple, absolutist statements.  In this case, I will not do so.

It isn't bad for RH.  It's frikken brilliant.  It's made them into a
very financially credible company.  Kudos to them.  They proved to the
market that free software is real and valuable, and they did that by
working around the structural bugs in the community.

My problems with that approach, though, are that it isn't sustainable,
long-term-rational, and, if carried out beyond necessity, it's
_unethical_engineering_.  In spite of the long hair and freakish
nature, I'm of the short-sleeves-pocket-protector crowd:  for better
or worse, it's part of my social function to put my foot down, from
time to time, and force an abort of the Challenger launch, if I'm at
all able to do so.

Outside of the universe of Dilbert, we engineers don't exploit
problems -- we fix them.  That's our job.  That's our legitimate
contribution to society.  An engineering business that suppresses the
impulse to fix things is a business that needs, uh, "correction".

They should be taking a two pronged approach.  Sure, bringing products
to market by offering the service of working around problems that
originate upstream in the flow of source -- but simultaneously working
to fix those problems upstream.  The unix vendors, especially IBM and
HP/C, should be giving them half a billion a year to do so -- with
that kind of funding, RH can tend to the commons a lot better than
those companies internal, closed-wall engineering teams.  (and, hey,
I'll take a year's sales commision on that :-)


	> RH wants OSS to succeed, but also wants there to be
	> opportunities to create proprietary lock-in.

That (lock-in as goal) is a bug.  They can't win that game.  They
don't do hw and they're too small.  It's bad for customers, and a lot
of customers aren't that stupid.  Their best medium-to-long-term bet
is to position themselves as the trans-corporate OS-group for the
companies who legitimately fight over those markets.  They ought to
focus on structuring the software pipeline so that vendor lock-in is
Not An Option for anyone.


	  > So in some cases they will proactively support the
	  > community, and in other cases they will let the community
	  > fend for itself and create a commercial product to fill
	  > the void.  I don't see a problem with that.


It's a treacherously slippery slope from the passive aggression of
"letting the community fend for itself" to the active aggression of 
"abusing the engineering community."

knowing nothing in life but to be legit,
-t