Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!ukma!tut.cis.ohio-state.edu!cs.utexas.edu!kelvin
From: kelvin@cs.utexas.edu (Kelvin Thompson)
Newsgroups: gnu.gcc
Subject: Re: Supposed intellectual property rights.
Summary: knowledge rights *do* exist
Message-ID: <136@yaxkin.cs.utexas.edu>
Date: 11 Jun 89 03:02:29 GMT
References: <8906100346.AA02699@sugar-bombs.ai.mit.edu>
Distribution: gnu
Organization: U. Texas CS Dept., Austin, Texas
Lines: 80

In article <8906100346.AA02699@sugar-bombs.ai.mit.edu>, rms@AI.MIT.EDU writes:
>
> [Quote by somebody else and followup: counter-examples from history.]
> 
> The text quoted above is misleading in another way as well: it speaks
> of "intellectual property rights" as if their existence and
> extent were generally accepted and uncontroversial.
> In fact, this is precisely what the controversy is about.

I believe I've heard RMS say he agrees in principle with patents and
copyrights as applied to their traditional targets -- physical
inventions and written works.  Under these conditions, it seems that
patents and copyrights are explicitly protecting "intellectual
property rights" -- they are allowing people to "own" ideas (or
expressions of ideas) for a period of time.

So the law clearly recognizes the *existence* of "intellectual property
rights"....and perhaps RMS does, too.  The question is under what
conditions do they exist, and how far do they extend.

> The Constitution itself might suggest a criterion: it says that the
> purpose of copyright is, "To promote the progress of science and
> useful arts."  In other words, if user-interface copyright actually
> impedes progress as many people in the field believe, it is
> unconstitutional.

Yes, an excellent criterion.  But I think disallowing all copyrights of
software would inhibit "the progress of science and art".  Let's take
two examples from fictional, parallel universes:

[1]  In this universe, light bulbs do not exist.  A few scattered
researchers have ideas about how to make light bulbs, but their
prototypes are dim, uneconomical, and unreliable.  One very smart
researcher who has made a lot of money on previous inventions decides
to make a major push toward discovering a workable lightbulb.  He
builds labs, obtains materials, hires workers; his organization spends
years testing all kinds of materials, voltages, currents, atmospheres,
bulb sizes, etc., etc.  Finally, after a lot of work and money, the
group develops an economic light bulb.  It would be trivial for others
to reverse engineer the bulb, but patent law protects the inventor --
he can get a fair return on his investment.  Without the patent
protection, he might never have made the investment, and a workable
lightbulb might have taken years longer to be developed.  With the
protection, science progressed faster.

[2]  In this universe, mouse-based computer/user interfaces are rather
awkward.  A computer company decides that, instead of hacking out yet
another awkward interface for its new computer, it will develop an
interface that is really *right*.  The company takes a lot of extra
time and money to explore several different ways a mouse-based
windowing system might work.  It runs real, scientific tests on users
to determine what methods objectively work best for a large group of
people, as opposed to what a couple of programmers like best.  Using
the results of the research, the company releases an interface that is
clearly the best around.  It would be trivial for another company to
reverse engineer the interface, but copyright law protects the
developer.  Without copyright protection......  [you get the idea].

[To reiterate: I stipulate that these stories are fictional.]

A couple of years ago, I attended a talk RMS gave here at the University
of Texas.  I agreed with most of his ideas, but I couldn't go for Total
Emancipation of All Software.  I think Universe #2 above is close enough
to reality that some copyrightability is best for all.

During the talk's Q&A segment, Stallman said that with a gun pointed to
his head, as an extreme compromise, he might be able to accept a
relatively short copyright period for software.  Say 3-5 years.  This
sounds ideal to me: it protects investment, but allows for public use
of the software before it is ancient history (I think one sofware year
is equivalent to about ten poem years; sort of like dogs and people.)

Just think, if 3-5 years were the law, then Apple's look-and-feel -- and
much of its ROMs -- would be going free about now.

[I apologize in advance if I've misrepresented anybody's opinions.]

-- 
-- Kelvin Thompson, Lone Rider of the Apocalypse
   kelvin@cs.utexas.edu  {...,uunet}!cs.utexas.edu!kelvin