On Wed, 13 Aug 2003, Keith Dunwoody wrote:
KD>Fedor Zuev wrote:
KD>>>Your definition seems to differ from standard usage.
>>
>> What is the standard usage?
>>
KD>I can't exactly define a "standard" usage, but in my experience most people
KD>would consider all binary executables as "software".
Even if it is a zip self-extracted archive?
Really, "my definition" is partly from a mathematical theory
of algorithms. Every algorithm should have input, output, be finite
(trivial for reallife) and monosemantic (trivial for digital
computers).
KD>However, I believe the difference between "software" and
KD>"documentation" is a bit of a red-herring. I have yet to see any
KD>convincing arguments why "documentation" (however one wants to
KD>define the term) should be allowed to be less free than any
KD>(other) software.
Why do you think that "different" necessarily means "less
free"? There may be different essential freedoms for the different
types of works, and, also, different non-freedoms under copyright
regime, which should be removed.
For example GFDL, unlike any free software licences,
specifically grant to user the rights for publicly display licenced
work and right to translate it. For the software, these rights not
exist as separate exclusive rights, or almost useless. But for the
documentation, as well as for artistic works these rights may be
essential, especially right of translation. Why you think that
grant these rights is "less free" than not grant it?
On the other hand in every country exist, in some form, a
statutory freedom to citation for the literary works. Not for
"computer programs as literary works" but for real literary works.
Modern society cannot exist anyway without citations. For literary
work you need not care to explicitly grant to reader the right to
make, include in his work and distribute a small excerptions from
your manuscript - reader has this right anyway, directly from a law.
You can safely assume that any "derivative work", for which you
grant licence is large enough, comparable by size with the original
work.
For literary works may not and, IMAO, should not exist the
freedom to "fix" the political or social views of the original
author, in the other words - freedom for censorship someone's else
work. Because this freedom is useless for any creative purpose. You
always can add you own opinion as commentary. And, anyway, you can
not grant such freedom in licence in advance - right to integrity
inseparable from the author. In advance author may only mark some
words he values as essential for "his or her honor or reputation".
KD>The arguments which I've heard basically boil down to: 1) The GNU
KD>manuals should be allowed into main, so we need to find some way
KD>to allow a license which would be considered non-free if applied
KD>to a program to be applied to the manuals.
KD>-- This is putting the cart before the horse. There are
KD>guidelines, and we should judge what is put into main by the
KD>guidelines, rather than judging the guidelines by what we want to
KD>put into main.
Maybe. But there also another element in the picture. For
GFDL. This is a not a random package from the random source with the
random licence. This is a licence from Stallman, the inventor of the
term "free software" and creator of the free software movement.
There is no reason to think that Stallman is an idiot and can not
see these semi-trivial arguments, presented here against GFDL.
Contrary, in the past Stallman so many times was right, in very
pragmatical sense of "right", whereas virtually everyone else was
wrong, so he shall be far more credible than random subscriber of
debian-legal. So, it may as well be seen as the test case for the
DFSG as for the GFDL.
Same for RFCs. It is the root of the Internet, which, in
order, is the material root of the many most valuable modern
freedoms.
