Subject: Re: the Free Software Movement in Industry
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Sat, 25 Aug 2001 18:16:00 +0900

>>>>> "Ian" == Ian Lance Taylor <ian@airs.com> writes:

    Ian> A one-shot prisoner's dilemma is actually resolvable

Only by making very strong assumptions.  If you read Hofstadter
closely, you'll see he's actually assuming that your thinking
determines others' behavior, not their thinking.  Assuming that
others' think like you is the standard assumption in game theory,
after all.  So Hofstadter has to be making a stronger one.

Either way, it's a pretty risky assumption.  Ask Yasser Arafat about
super-rational opponents.

    Ian> In any case, a one-shot prisoner's dilemma is rare.

Well, yes, that does mean the outcome will be different.  That's more
or less my point.  To change the outcome, you need to change the game.
Repeat it, add strategic options, lobotomize the players (that's how
Axelrod gets his result).

However, the kinds of real-world situations we're looking at are close
enough to one-shot prisoners' dilemma to make the analogy something to
worry about.

    Ian> Anyhow, I don't think writing everything as free software is
    Ian> a prisoner's dilemma, because it's not obvious that, e.g.,
    Ian> everybody is better off if telco billing system is free
    Ian> software.

Oh, I think it's close enough to true to stipulate it.

It's the other parts of the game that we need to worry about.  Can we
set up the rules for bidding on projects so that the principal does
not have the incentive to choose the low bidder, rather than the
author of software?  (BTW, I see no justification for the presumption
that free software authors know more about maintaining their package
than a competent maintenance specialist.  Eg, set Cygnus Solutions
(RIP) against the GNU Project.  Who would you choose to maintain your
purchased GNU package?  Cygnus, every time, for me.  Even if they
charged more.  DJB vs. Crynwr for your qmail system?  A no-brainer.)

Can we get enough repetition in a market where the principals have a
habit of going away because it's not fun any more?  And we're
unwilling to tie them to their last product by giving them IP in it?

How do you deal with the Bill Gates who sees a way to make big gains
by taking advantage of the current situation and is not about to agree
to rules constraining his freedom?

Etc, etc.  Tom's proposal and responses simply don't address these.

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
_________________  _________________  _________________  _________________
What are those straight lines for?  "XEmacs rules."