Paul Fisher wrote:
>
> Alex Nicolaou <[EMAIL PROTECTED]> writes:
> > Sun only requires compatibility for code which will be part of the
> > distribution that Sun releases.
>
> And these `compatibility' terms are non-free. The FSF will never
> distribute your modifications in a non-free fashion; Sun will.
I'm not sure what you mean by 'non-free' in this context. It is true
that the compatibility terms are not free in the sense that they
restrict what modifications you are permitted to make. However, they do
not bear a dollar cost; you return your modifications to Sun and they
are released (or not) in the next distribution. This is not much
different from giving code to an open-source project in the form of a
patch and waiting to see if it makes it into the next tarball, although
at least in the open-source case you can distribute your own
distribution. Below I'll argue that that isn't truly an option...
> > Requiring these royalties doesn't make the software 'not free' in
> > the FSF sense.
>
> Requiring royalties *does* make the software non-free, using either
> the Free Software Foundation's definition or the Debian/Open Source
> Free Software Guidelines definition.
>
> You might want to read ``Why Software Should Not Have Owners'' (off of
> the FSF philosophy page) for the ethical background on why royalties
> make a piece of software non-free.
I've read this page, but I don't think it has anything to do with money.
I think the following paragraphs cover the intent of the page clearly:
What does society need? It needs information that is
truly available to its citizens---for example, programs that people
can read, fix, adapt, and improve, not just operate. But
what software owners typically deliver is a black box that we
can't study or change.
Society also needs freedom. When a program has an owner, the
users lose freedom to control part of their own lives.
And above all society needs to encourage the spirit of voluntary
cooperation in its citizens. When software owners tell
us that helping our neighbors in a natural way is ``piracy'', they
pollute our society's civic spirit.
This is why we say that free software is a matter of freedom, not
price.
(http://www.gnu.org/philosophy/why-free.html)
Well, let's see - the Java 2 license *does* let you read, fix, adapt and
improve the software. It isn't a 'black box' that you can't study or
change. There is a lack of freedom in the way that Sun restricts you
from returning changes to your fellow man - via Sun - but what seems to
be getting missed in this argument is that *free software suffers from
this same lack of freedom*. In a sense, free software does this in a
much more sinister way, since it restricts your practical ability to
produce your own distributions but never openly admits to the norms that
make such distribution taboo. At an ideological level, free software
does not have this problem, but at a practical level it does, because
everyone wants to avoid `forking' and wasting resources. Look at the
teeth-gnashing over whether Japhar/Classpath compete with
Kaffe/Klasses.zip - clearly both groups have a certain amount of
discomfort that essentially parallel technical efforts exist, and exist
partly because Japhar/Classpath are LGPL instead of GPL!
I think that Eric Raymond makes the case very clearly in his paper
"Homesteading the Noosphere":
In practice, however, such `forking' almost never happens. Splits
in major projects have been rare, and always accompanied by
re-labeling and a large volume of public self-justification. It
is clear that, in such cases as the GNU Emacs/XEmacs split, or the
gcc/egcs split, or the various fissionings of the BSD splinter
groups, that the splitters felt they were going against a fairly
powerful community norm.
In fact (and in contradiction to the anyone-can-hack-anything
consensus theory) the open-source culture has an elaborate
but largely unadmitted set of ownership customs. These customs
regulate who can modify software, the circumstances under
which it can be modified, and (especially) who has the right to
redistribute modified versions back to the community.
(from
http://www.tuxedo.org/~esr/writings/homesteading/homesteading.html)
On your other point re: open source definition/Debian license, I agree
that the software doesn't meet the open source definition
(http://www.opensource.org/osd.html) on several counts, namely 1, 4, 5,
6, 7, 8. #1 is specifically about royalties. I haven't read anything on
the www.gnu.org site to suggest a similar prohibition, but on the
contrary have read on the FSF site that the opensource people are
wrong-headed and are too busy sugar-coating the grand ethical benefits
of open-source software with business considerations instead of
emphasising the inherent value of running free software as an end unto
itself. This criticism is returned by the open-source site which says
that open-source software doesn't need "philosophical tub-thumping".
What a shame we can't spawn such open dissention within the ranks of
Microsoft - wouldn't it be great to read pages by the NT team cutting up
Windows 95 and vice versa?
But I'm getting away from my point. If the FSF really objects to
royalties they'd better update their pages to be a little less
ambiguous. I still claim that the main reason the Java 2 license isn't
free is because of the way your mods have to be returned to the public -
via Sun - and because of the fact that your mods are restricted to Sun's
definition of Java. While these two restrictions aren't particularly
palatable, especially since Sun is bound to be slow releasing updated
versions of the code with your patches in it, they aren't really much
different from the practical reality of participating in open-source
development. Only those with keys to the CVS tree truly have 'hack as
you will' access, and the fact that the CVS heirarchy is protected from
the random hacker accessing it doesn't strike me as any different than
Sun's legalese-laden policies. Certainly it's no different at an
ethics/morality level.
alex
