Subject: Re: Open letter to those who believe in a right to free software
From: Ian Lance Taylor <>
Date: 4 Nov 1999 15:53:41 -0500

   Date: Thu, 04 Nov 1999 12:26:40 -0800
   From: "Tim O'Reilly" <>

   > The history of free software to date has been highly contingent.  If
   > there were no RMS, there would not have been anybody else.  If there
   > were no Linus Torvalds, there would not have been anybody else.

   I am sorry, but this is just contrary to the facts.  Without denigrating
   the contributions of either Richard or Linus, the sharing of source code
   and giving it away to a wider community far precedes either of them.  

I didn't say that they invented giving away source code.  I didn't
mean to imply that, nor did I mean to denigrate the many other people
who have created free software, nor did I mean to present any sort of
revisionist history.

What I said was that the history of free software has been highly

By that I mean that it could have followed significantly different
paths, paths so different that the current software world would look
significantly different than it does today.  Moreover, I mean that the
choices between the different paths it could have followed have been
significantly affected by individual choices.

For example, you mentioned Usenet.  Usenet was a great invention, but
I would argue that it was in some sense a likely one.  As evidence I
present the more or less independent inventions of bulletin boards and
Fidonet.  If Usenet had not been invented at Duke and UNC, I believe
that something similar would have been invented somewhere else.

   Success has a thousand fathers.  We are all part of a great river with
   ten thousand sources.  Trying to trace it back to one common ancestor
   just misses the boat.  The same is true in looking at the early history
   of the Internet:  there was no "father of the internet" but a tribe of
   people contributing ideas to each other, some more important than
   others, but none able to stand alone.

   The fact is that without information sharing there can be no progress in
   an intellectual community.  The amount of sharing can speed up or retard
   that progress.  That's all we know.  We're all still struggling to
   understand the exact dynamics.

I agree.  What I'm saying is that the dynamics of free software are
such that single individuals can affect it significantly, making a
statistical treatment very difficult.

The same is true of science, although the time scale is not as
compressed as in computer software today.