rms@AI.MIT.EDU (06/03/89)
Second, and this is something most people don't think about, GNU
software is not "free". There is no money involved in getting it, but
it is proprietary software as long as there are restrictions placed on
its use and redistribution. Calling it "Free" is an interesting use
of the language.
1. I have defined "free software" as software which everyone has the
freedom to run, redistribute and change. This is what I have meant,
every time I have used the phrase since I started the GNU project.
2. I believe an acceptable definition for "proprietary software" is
software which people are forbidden to run without individual
permission from a specific party (usually offered in exchange for
money).
3. By these definitions, GNU software is free, and is not proprietary.
4. It is true that there are restrictions on the use of GNU software.
These restrictions prevent other people from imposing restrictions on
the use of GNU software. (I have described and explained these
restrictions, and their motivation, so many times that I am suprised
anyone can find them astonishing or paradoxical.)
5. By contrast, public-domain software (which GNU is not) has no
restriction imposed by the author. However, as a result, any
redistributor can impose restrictions on the people who receive copies
via that redistributor. Thus, we have in practice a paradox deeper
than the simple one spaf proposes: that software can be in the public
domain, and at the same time proprietary.
6. For example, I have used a machine which has X Windows on it, as a
proprietary program. (X Windows is not public domain, but its copying
conditions are effectively public domain.) It does not come with
source code, and changes have made some internal interfaces incompatible
so that the source code from MIT is not useful for fixing the bugs in
the object code.
7. Therefore, no matter what I do, someone is going to be restricted in
some way. My choice can only affect what kinds of restrictions are
possible. I don't believe that a restriction against adding restrictions
is wrong in the same way that a restriction against sharing and changing
software is wrong.
8. A good analogy to understand this is the concept of "inalienable
rights". I have tried to create an inalienable right to share and
change GNU software.
For example, you might say that your freedom is limited because you
don't have the ability to sell yourself into slavery (not such as the
rest of society will recognize as binding). However, if people could
do this, and it started to become commonplace, many people would have
much less freedom than they do now.
fozzard@boulder.Colorado.EDU (Richard Fozzard) (06/06/89)
I want to take this opportunity to thank RMS for his reasoned and rational postings about the FSF philosophy. They are a breath of fresh air compared with the impassioned rhetoric of some of his disciples. This doesn't mean I agree with everything - hardly - but that my faith in the goals of FSF is being restored. ======================================================================== Richard Fozzard "Serendipity empowers" University of Colorado fozzard@boulder.colorado.edu (303)492-8136 or 444-3168