Permission to use portions of the recent GNU Emacs Manual
Robert J. Chassell
bob at rattlesnake.com
Sun Dec 12 14:43:59 EST 2004
> To enable the FSF to be a `do what we do organization' rather than a
> `don't do what we do, do what we say' organization.
a) what precise purpose does the GFDL achieve for the GNU Emacs manual
that is not achieved by it being under the GPL (your explanation is
mostly about the contrast to Public Domain and/or BSD)?
To be a `do what we do, not a do what we say' organization.
b) is the difference this makes worth the trouble it causes?
I think so. Certainly, over the past few years lawyers have become
very important in the question of software and other freedoms. As the
cost of manufacturing information, which is called `copying', comes
down, the information's legal status becomes more and more important.
A long time ago, I heard someone say that the issue was whether anyone
could create a program using an interpreted language like Emacs Lisp,
or whether the program would be too slow. That is not the issue now.
Many programmers think of the trouble licenses cause to other
programmers. That is to be expected, since programmers write code and
But the issue here is the interface between certain professionals in
one industry and everyone else. Moreover, industry also develops
hardware that combined with software have enabled copying to become
less and less expensive. This means that when you show how the
interface is supposed to work, you need to think about all the people
who are not programmers and who do not know anything about computers
Nanotechnological assemblers -- rapidly reproducing Von Neumann
machines that also copy other things -- would reduce the cost of
manufacturing copies to zero. But we are never going to see a
reduction to zero in the cost of attention getting, that is to say, of
marketing and advertising. That is because getting attention is a
zero-sum operation, like getting status. It is not like science-based
manufacturing, which is positive sum.
This means that an appropriate license will continue to be relevant.
c) in particular: is it a good idea to split a free project into two
parts with incompatible licences in a manner that makes it only
possible for the copyright holder to sensibly maintain the exchange of
material across the rift? Do we really want to demonstrate how to do
such a thing in large scale?
If people on one of the two projects insist on using a license that
hurts people, especially the vast majority, who are non-programmers,
then their project should suffer.
So far, I have not seen any remarks that speak to the points. Among
others, I would like to see an argument favoring a `Creative Commons'
license that forbids any but gratis distribution, except for its
controlling monopolist. Why would that be better for you and me and
others? I think it is worse.
I would like to see an argument that says that says that FSF need not
be consistent; that it should become a `do not do what I do, do what I
say' organization. I think the FSF should be consistent. Perhaps I
I would like to see, but do not expect to see, an argument that says
that many programmers who work on GNU projects are really good at
selling non-programing business patterns to people are not programmers
and who know nothing about computers and software.
d) is it a good idea to change a large body of free software (like
the GNU Emacs manual) to a different licence when it is well-known
that substantial forks exist for which no licence change is
Why is it not possible? Are you suggesting that the XEmacs people
lack proper papers for contributions? If so, this means they cannot
ever ensure that the lawyers of a hostile organization with plenty of
money will tell the organization's leaders that the software is legal.
That is what I think you are suggesting.
This means that the XEmacs project is rhetorically ineffective. Since
I doubt that many people on it are concerned about business patterns,
politics and the like, I do not think this issue is salient to them.
I think they are mainly concerned with programming and wish the wider
world would go away.
In any event, the Lucid people were warned about this years ago and
continued on their way. As for FSF: why should a project favoring
software freedom, the GNU project, change its licenses to hurt its
I think that a good reason to go back on the license would be that FSF
need not be consistent. But I do not believe that reason and have
never heard an argument that claims that.
Certainly, licenses are irrelevant to people who live in small
communities. In a small community, social pressure and personal
knowledge works well. 25 years ago, Unix and Lisp programmers lived
in such a community or many of them did. I think many programmers
wish they still lived in such a community.
But the wider world has intervened. Programming and its machines have
become successful. Even my grand-neice uses a computer. We do not
live in a small and isolated community any more.
As for living in a bigger world: one could argue that laws and
licenses are irrelevant since most of the world is `extra-legal'.
Only regions like Western Europe and a few others are more or less
`law abiding'. In much of the world, any organization with more than
a billion US dollars or so per year to spend can corrupt enough
officials for it to continue to do what it has done before.
But again, I do not think that the rule of rich, greedy individuals is
better than the rule of law. I have heard claims that rich, greedy
individuals do make better rulers, and that their right to hinder me
is good for others than me; but I have never welcomed that.
Moreover, there are good arguments -- indeed I have made them -- that
failure is an intrinsic problem of hierarchies that exclude change.
(Failure may not occur in the short term, that is to say, in less than
a couple of human generations.) Only when competitors are equally
weak can such hierarchies last for thousands of years.
So, to answer your questions again:
a) the precise purpose of the GNU Free Documentation License is to
offer a better alternative to greedy publishers, a better way to
protect their investment in attention getting. The FSF puts its
own documentation under the GFDL because its leaders believe that
programmers, at least those involved in the GNU project, are poor
at being `don't do what we do, do what we say' salesmen to people
who are not programmers.
b) the difference between a license that insists that commercial
organizations fail in their purpose and a license that enables
their purpose is vast. No one claims that companies that are
unable recover costs will survive against companies that are able
to revover costs.
c) a project that hinders another in fulfilling its goal cannot expect
to be supported by the second, except as charity.
d) as an organization comes to realize that its primary business is
politics in a deep sense -- that of changing business patterns --
and that the people in the organization are poor at politics with
strangers who know nothing of computers and that competitors are
out to destroy it, it makes more sense to offer licenses such as
the GFDL as an alternative to licenses such as the `Creative
Commons license with a commercial restriction'.
Robert J. Chassell
bob at rattlesnake.com GnuPG Key ID: 004B4AC8
More information about the XEmacs-Beta