🚀 go-pugleaf

In article <37s1ci$98q@nntp1.u.washington.edu> tzs@u.washington.edu (Tim Smith) writes:

   Craig Burley <burley@gnu.ai.mit.edu> wrote:
   >I would love to know the answers.  (I do hope you showed your professor
   >more than just my post -- like the GPL itself, the FSF's posted
   >interpretations, and such.)

   We seem to be going in circles here.  A couple of posts ago, I wrote:
   <Put another way, the way one becomes bound to GPL is by doing something
   <that would infringe the copyright of some GPL'ed code if one did not have
   <permission to do that thing.

   You seemed to agree with that, for you responded:
   >Good so far.

Good so far.

   The point where I disagree with you, and where my professor agrees with
   me is over whether or not what was distributed as part of RIPEM infringed
   the copyright of gmp.  We say no, and hence we say that GPL and the FSF's
   interpretations of GPL are not relevant here.

   As far as I or my professor can tell, there is no concept of implicit
   distribution in copyright law.  If the *actual* *files* distributed
   as part of RIPEM do not contain copyrighted parts of gmp, then their
   distribution is not an infringement.

Yes, indeed, you've posted a good summary of a major aspect of the
discussion, and yes I do understand what you and your professor
are saying you believe.  (In fact, I believe I've understand that for
a very long time.)

The other aspect of the discussion that you seem to be missing, or
not responding to, is this:

Are you going to continue to act as if the _legal_ opinion you and
your professor hold is so 100% absolutely certain (even though we have
evidence that there are practicing IP lawyers who at least _claim_
to disagree) that anyone who claims to disagree must have some other
agenda for enforcing their license restrictions, and thus (for example)
continue to claim or imply that the FSF has, is, or will claim an
interface copyright (however "virtual") or might someday claim a
copyright on any and all code written in a language implemented by
FSF code?

The only reason I entered this thread was because you said the FSF's
reasoning regarding RIPEM could be applied to code written in
languages like GCC and EMACS extension language.  (You didn't use
those example, but they apply, since they implement languages, are
GPL'ed, and are the only implementations of those languages -- the
conditions being discussed.)  The only way you can do that (assuming
you have a reasonable amount of ethics/honesty) is if you assume
that

-  The FSF knows its claim that a work not explicitly including
   GPL'ed code can be distributed without falling under the GPL,
   even if it is clearly designed to be automatically joined up
   with GPL'ed code to form a result that itself cannot be
   distributed under the GPL

AND that

-  The FSF nevertheless continues to make its false claims in public

AND that

-  The FSF has some other agenda it plans to use should someone
   "violate" the GPL a la gmp/RIPEM and force the FSF to take the
   violator to court

AND that

-  The FSF would, in court, defy its earlier public claims that it
   does not assert "interface" or "language" copyrights and assert
   them (in at least some form) to make its case

Whether you admit it or not, you have made several posts in the recent
past that have impugned motives to the FSF.  And you seem to have
arrived at these suspicions via a very tortured road of so-called
logic, starting with the assumption that just because you and your
professor don't believe something about copyright law, the FSF
must not believe it either, and therefore any claims it makes about
what can and cannot be done with the GPL must necessarily _not_
be based on that something -- so any other basis can be invented
wholesale by you, even if that basis has been explicitly disclaimed
by the FSF.

Meanwhile, even _I_ agree that you and your professor might be right,
but I tend to not suspect the FSF of all of the above (obviously).

My view is simple, and follows a rather simple path of logic (though,
as usual, it is lengthy in verbage because of the defensive way
pro-FSF posts must be worded to reduce excessive flamage):


"The FSF appears and claims to believe that there are cases where
distributing a work that is apparently absent any GPL'ed code might
nevertheless, under certain circumstances that are likely to be
self-evident for any given case, considered by copyright law to
be distributing a modified version of one or more GPL'ed works (which
I call a 'phantom distribution' claim).

"The FSF might be wrong, but it might be right.  Even if I was an
expert on intellectual property law and was 100% sure it was wrong,
that would not make me an expert on what the FSF believes, or on
whether what it believes is the same as what it claims.

"Therefore, to attempt to determine the FSF's beliefs, I must look not
only at its public statements, but at its actions, especially
regarding any sinister motives I might believe it has.

"If the FSF did not really believe in its 'phantom distribution' claim,
what agenda might it have to attempt to enforce the GPL should an
important court case arise?

"One claim is that it would assert an interface copyright of some sort.
This is something a 'software-hoarding corporation' might do, and
it certainly would restrict the ability of people to write their
own non-GPL'ed GNU-compatible software from scratch, so certainly
it would be a very insiduous and hypocritical thing for the FSF to
plan to do.

"However, the FSF has explicitly disclaimed interface copyright, and
furthermore is a corporate member of an organization (the LPF) that
is fighting things like interface copyrights.  Given the generally
wise legal advice not to speak in advance about a possible future
case, it is hard to imagine that publically disclaiming interface
copyright has any advantage for the FSF if it plans to use interface
copyright in attempting to enforce the GPL in the future.  In fact,
such a public disclaimer is much more indicative of an organization
willing to box itself into a corner as far as how successful it will
be in enforcing the GPL, so as to make its own principles clear and
unassailable.

"Further, the FSF has allowed the interfaces of some of its products
to be explicitly and exactly copied, without objection (and in one
case with actual _encouragement_), which suggests it is not engaging
in a quiet campaign to protect its interfaces as property, something
that might be important in this new and largely untested area of IP law.

"Similarly, using a 'language copyright' claim to enforce the GPL
in other cases, cases where the work contains no GPL'ed code but is
written in a language implemented only by GPL'ed code, seems unlikely
since the FSF has explicitly disclaimed such a claim, and further
has allowed such works to be distributed in the past -- works, based
on GCC or EMACS' extension language as unique languages, distributed
but not under the GPL.

"Therefore, it seems simple and straightforward enough to conclude
that the FSF really does believe in the 'phantom distribution' claim
(as I call it), or at least wants people to believe in it, but in
any case that it does not have any other identifiable agenda for
enforcing the GPL, or at least not one involving claiming interface
or language copyrights.

"As a result, I see no reason to publically impugn the FSF, or the
FSF's motives, either explicitly (by saying the FSF has claimed
an 'interface copyright') or implicitly (by saying the FSF's
arguments regarding the GPL are applicable to a 'language copyright'
claim).  To do so would be to publically impugn the motives of an
organization that is doing good work in the computing community,
in that it distributes useful software for free.  And, more than
likely, to do so would probably tell people more about the sort
of person I am than the sort of organization the FSF is."

--

James Craig Burley, Software Craftsperson    burley@gnu.ai.mit.edu