Subject: Re: the GCC steering committee
From: (Kragen Sitaker)
Date: Fri, 15 Mar 2002 20:35:20 -0500 (EST)

Jonathan Shapiro writes:
> How many forks of the RedHat tree have happened over the years? How
> many has the market actually paid any attention to?

Caldera, Mandrake, and Red Flag are three off the top of my head.
Both Caldera and Mandrake eventually built their own distributions
from scratch, but both started out as modified Red Hat distributions.
Red Flag is still a modified Red Hat distribution, but it's rather
young yet.  The market has paid attention to these three.

There's also WireX's Immunix.

I don't follow Linux distributions closely, so for all I know, there
could be hundreds of them.

> If I forked the RedHat distribution tomorrow, would *anybody* --
> even those of you on this list -- give a damn? Heck. Would anybody
> even *notice*?

Evidence suggests it would depend on how well you publicized it and
how well it worked.  The history with EROS suggests that you would do
a good job of publicizing it, and people would get very excited about
it, but it would take at least ten years to work well enough to be
useful, so nobody would notice after a while.

Discussing "forks of a distribution" is a little weird, though.  Every
Linux distribution is a "fork of the Red Hat distribution" to the same
extent that egcs was a fork of gcc or XEmacs is a fork of GNU Emacs
--- they include almost all the source code in Red Hat's distribution,
even if they aren't really based on Red Hat.

> Any idiot can fork a tree. The power to succeed in doing so is a
> function of marketing and PR, not the ability to fork.

If you don't have the ability to fork, you can't succeed in doing so.

> This point is critical. It is the fundamental reason why a VC
> investor can safely invest in open source -- if the successful
> investee maintains leadership, they maintain lock.

Yes, you are right, and Cygnus has done exactly that for many years.
Cyclic did the same thing until their full-time employee decided he
could make more money working for Cygnus.

For reasons that have been discussed to death on this list, the annual
ROI on free software ("open source") development is not likely to be
in the 1000% range typical VCs expect.  VCs investing in free software
businesses are going to have to content themselves with much lower
ROIs, or, more likely, they're going to turn the company into a
different kind of business in hopes of increasing its ROI.

<>       Kragen Sitaker     <>
Perilous to all of us are the devices of an art deeper than we possess
       -- Gandalf the White [J.R.R. Tolkien, "The Two Towers", Bk 3, Ch. XI]