Subject: Re: the walls have ears
From: "Stephen J. Turnbull" <>
Date: Mon, 31 May 1999 12:05:41 +0900 (JST)

>>>>> "kms" == Karsten M Self <> writes:

    kms> As I understand, the effect was similar to what might be
    kms> possible under, say, the BSD or Artistic License, where a
    kms> third party is allowed to spawn a proprietary development
    kms> tree, subject to some limitations, but not providing the four
    kms> software freedoms enumerated by RMS.

    kms> RMS saw this, 


    kms> and it was not good.

This is called "assuming the conclusion."  At least some of us
consider this to be the point at issue, and it is _not_ moot.

Until[1] somebody works out an appropriate model of industry dynamics,
I will consider it theoretically possible that BSD-style licensing
will result in greater production and greater use of free software in
absolute terms (figuring out how to measure this, even in theoretical
terms, will be difficult) than does GPL.

If this is so, it will be impossible to theoretically rule out BSD as
a candidate for "best free license" in social terms unless you are
also willing to rule out utilitarianism.[2] (On this thread I suspect
Jean Camp has such leanings, and probably also Richard Stallman,
although I don't know that either would completely reject
utilitarianism.)  And since BSD probably leads to more profit, too
(Brock Lynn's boisterous advocacy notwithstanding), that would make
the BSD pole very attractive[3] to many FSBers, I suspect, although
many might prefer an Aladdin-style dual licensing program, since using
GPL as the free public license gives you an effective (although
temporary) monopoly on proprietary improvements.

[1]  Yes, until.  I for one am working on this.  No, that doesn't
guarantee that my current beliefs will be justified by such a model
;-)  It is possible to do scientific economics.  :-)

[2]  Ie, the assumption that a change that makes everyone better off
evaluated by their own preferences (an improvement in economic
efficiency according to the strict definition) is a social
improvement.  Strictly speaking utilitarianism assumes that utility is
interpersonally commeasurable, but that is strong enough that it
becomes a straw man.  Without interpersonal commeasurability, the
efficiency principle is the closest approximation you can get to
utilitarianism (as far as I know; it's a huge literature and I'm not a 
specialist), and the policy conclusions don't differ.

[3]  Of course with modifications, for reasons that Russ Nelson and
Brian Behlendorf have been discussing.

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 two straight lines for?  "Free software rules."