[ On Tue, September 22, 1998 at 12:19:54 (-0700), Bill Studenmund wrote: ]
> Subject: Re: More licensing flames...
>
> On Tue, 22 Sep 1998, Greg A. Woods wrote:
> 
> > [ On , September 21, 1998 at 23:00:40 (-0700), Chris G. Demetriou wrote: ]
> > > Subject: Re: More licensing flames...
> > >
> > > woods@most.weird.com (Greg A. Woods) writes:
> > > > The Sleepycat copyright license on their DB-2.x goes one step further
> > > > and contains terms that some might describe as "viral" -- i.e. they
> > > > attempt to force people who distribute software linked against DB-2.x to
> > > > also be freely re-distributable in source form.  I.e. the DB-2.x
> > > > copyright license is very much more like the GNU LGPL than the GPL.
> > > 
> > > Uh, do you have those reversed?
> > 
> > Nope.
> 
> Uhm, as I understand all of this, the sleepycat license is more like GPL
> than LGPL.

Ah, well, I guess it depends on wether you agree that the GPL is a worse
"virus" than the LGPL, or not.

The Sleepycat DB 2.x copyright license is indeed clearly more
restrictive than the GNU LGPL (i.e. the latter allows object-only
distribution of the base program where the former attempts to force full
source distribution).  If indeed either would be found to be legal
restrictions under the Copyright Law.

> I don't think that anyone has been arguing that linking w/ (L)GPL'd stuff
> means you loose the copyright on your code. If they were, I missed it. 
> 
> The thing though is that (L)GPL has a restrictive LICENSE on how you can
> distribute such code.

Whenever I talk about copyright licenses I talk *only* about
distribution or other acts that would contravene copyright law.

> > Besides, it refers to using "parts" of the Program, and implies
> > derrivation.  Most times a library, esp. a specific library such as DB,
> > is used in an application the entire content of the "Program" is
> > included with the distribution of the application that uses it.  The GPL
> > section 2 attempts to "control the distribution of derivative or
> > collective works based on the Program".  Including the "Program" as a
> > library which a larger proprietary application uses does not form a work
> > "based on the Program", but rather one based on the application owned
> > entirely by proprietary software vendor.  I.e. the emphasis is extremely
> > important.
> 
> Actually, sticking the program inside another program would be covered by
> section 2 as I understand it. That's what GPL is trying to prevent. You're
> right that the proprietary software house would still own all the other
> parts. That's not the problem, though. They couldn't legally redistribute
> the program if it included parts which were from GPL'd code UNLESS they
> made the source for the whole thing available. 

Ah, no, that's exactly the opposite of what I've been trying to say.

Remember, collective works are explicitly permitted under law to be
covered by a copyright that is totally separate from the copyright that
may be on the original component works.

In the absence of case law that shows an interpretation stating
otherwise, it is my understanding that copyright law, at least in
Canada, will not allow the copyright on a component of a collection to
affect the other components in the collection.  Only the copyright of
the entire collection may be affected, *and* the only affect it could
likely have would be to totally prevent distribution.  It is highly
unlikely that the copyright license on a component of a collection would
be able to force someone to reveal what is in effect a trade secret
(eg. some other component in the collection), even if such a license
made claims that doing so would permit legal distribution of the
collection.

And it gets even more complicated....

> It is though perfectly possible for the license to dictate how the
> collection is distributed, though. 

Yes, but not in the way some claim the GNU GPL or the GNU LGPL (or
indeed the Sleepycat DB copyright license) attempt to do.  Traditionally
(eg. in the literary world) for a work to be include in another
collection a one-time license must be granted to the author of the
collection.  However there's little or no "tradition" for the case where
a freely re-distributable software component is linked into another
program, possibly one that's protected as a trade secret and where only
binary versions of the resulting application are distributed with an
"All rights reserved" collective copyright.

It is my understanding that under copyright law there's little or no
distinction between a program being linked with a library and having
that library distributed in a separate "file" on the same "media".  Both
will likely be viewed simply as "collective works".  Although the GPL
attempts to make this distinction on behalf of the law, I'm don't think
it would wash clean in the legal arena.

Take for example the case of a proprietary application system that uses
some interpreter that's distributed under the GNU GPL copyright license,
and which encrypts or otherwise obfuscates and token compiles the
interpreted code so that it cannot be re-engineered.  The GPL on its own
will view this as OK because the interpreter is being distributed as a
part of the collection of files that makes up the total application.
One might even argue that the use of a GNU GPL'ed interpreter is not a
necessity -- it is merely economic convenience to the software vendor
(eg. in the case where there are alternate interpreters available).  If
instead the interpreter is embedded in some part of the application but
the core of the application is still effectively the interpreted code
that the interpreter runs, (eg. a proprietary re-implementation of
something like Tk is linked with a GPL'ed interpreter such as guile
that's modified to read, and the application is actually the obfuscated
and tokenized scheme code).  Someone attempting to defend the GPL to its
limits may try to claim that not only the Tk-like thing should be
distributed with source code, but so should the obfuscated scheme
portion.  However the legal system will likely view the entire thing as
a collective work and so long as the vendor makes overtures to offer the
source code of the interpreter itself to their customers, it will find
that the vendor has upheld the GNU GPL copyright on the interpreter just
as they would in the first case where the interpreter remains standing
on its own.

What about a proprietary source code control system that merely invokes
a GNU GPL'ed program (eg. GNU Diff, or GNU RCS, or some such)?

What about an operating system that relies heavily on GPL'ed components,
such as the way NetBSD relies heavily on the GNU C Compiler?

What I'm trying to get at here is how one would make the distinction in
the eyes of the law between actually linking library code into an
application and simply accompanying an application with some other
separate program.  How closely does one have to tie GNU GPL'ed code to
an application before it is considered to be the same work vs. a
separate work in a collection?

I contend that the "closeness" of GNU GPL'ed code is irrelevant to the
law and that the GNU GPL virus is unenforceable.

Indeed I contend that the stated intent of the GNU GPL copyright license
is *not* to prevent distribution as a component in a collective work,
*regardless* of how closely tied the GPL'ed code is to the proprietary
code.

It is in fact irrelevant how the creators and users of the GNU GPL
attempt to interpret it, or how I interpret it -- the only
interpretation of importance would be that of a court of law in the
relevant jurisdiction (and from a third party's perspective enough case
law to build up a consistent interpretation that would allow the third
party's legal counsel to give reasonable determination of how the law
might affect that third party).

There are also "political" and economic issues that affect how GNU GPL
code might be defended in the real world.

So far as I know the law isn't anywhere sophisticated enough to make
some of the fine distinctions that would be required in a case against a
vendor using a GPL'ed library.  Any attempt to defend such a case
reliably would likely require many months of preliminary legal research
and probably weeks of explanation in front of some judge.  The expense
to a defending organization such as the FSF, would be very high, which
would effectively dictate that they would only pursue a big fish.
There's also a related problem because courts are under pressures due to
over-load, both in the USA and Canada, and if the damages claimed are
not sufficiently high to justify the court time and costs that would be
required for such a complex case, it may be simply dismissed as
irrelevant.  The politics and economics of defending free software are
not simple.

In the end though even if the author of the "Collective work" (i.e. the
proprietary application) was found to have violated the copyright of
some GPL'ed component the only award would be "reasonable damages".  The
courts would most likely side with a company trying to make an honest
living, not with the FSF, and if so the damages would only be to pay the
FSF some reasonable amount for the use of the GPL'ed component.  It
would be very unlikely for a judge in Canada or the USA to force an
otherwise law abiding company to convert their trade secrets into freely
available code and thus totally destroy an asset of the company which
may in turn totally destroy the company.  We won't really know though
until there have been a significant number of relevant cases in various
jurisdictions.

If anyone an come up with a real-life example of how a GNU GPL style
copyright actually resulted in the "freeing" of some other work that
accompanied the already "free" one as a part of a collection, I'd be
more than great-full to hear of it.  (An absurd and likely impossible
example might be where inclusion of a public domain poem in a book
caused all the other poems to enter into the public domain.)

My example was intended to draw a parallel to well understood parts of
the copyright law and its application (i.e. parts where there is lots of
case law in many jurisdictions to give a better interpretations and
predictions about un-heard cases):

> > For example consider what might happen if a poem that's got a copyright
> > similar to the GNU GPL is incorporated into a book or magazine that is
> > covered by an "All rights reserved" copyright.  This book or magazine
> > would not be "based" on the poem.  The author of the book or magazine is
> > free to sell this collective work for a profit.  However I or anyone
> > else can freely distribute copies of the pages on which that poem
> > appears, but we cannot distribute copies of the entire book or magazine,
> > and we cannot necessarily distribute copies of any other poems or
> > articles in the book regardless of whether those other parts were
> > written by the author of the collection or by third party authors.  The
> > authors of any other works that are included in this book or magazine
> > may still be free to distribute copies of the pages on which their works
> > appear (though this will depend on the exact terms of the one-time
> > license they granted to the author of the collection).  In fact if I'm
> > remembering right we can even copy any "small" preamble that the author
> > of the collection may have included at the beginning of the poem because
> > that would be considered "fair use" (so long as I play by the "fair use"
> > rules, of course, and acknowledge the source of the preamble).
> 
> Actually, I think you couldn't copy THE PAGES from the book, but you could
> copy the poem. I.E. you could type it into another book if you gave
> credit.

Actually, yes, you or I can legally distribute copies of the exact pages
of the book or magazine.  That's what the "fair use" rules are all
about.  I'm assuming the actual pages will also reference the GNU GPL
style copyright license and mention the original author, which will mean
that in making a physical copy we are indeed meeting the original
copyright license terms.  We will, of course, also be giving credit to
the collection's copyright owner, esp. if we're using the preamble,
since that's required to be considered "fair use".

If we simply copy the words then we need not give credit to where we
found them (i.e. the collection in which we happened upon them, since
they are freely, and in theory anonymously, re-distributable from the
original author) -- we would only need to credit the original author.

Of course if the poem in the collection is actually in the public domain
then one could do anything at all with it.  You could even claim it as
your own (so long as the fact you found it in some particular collection
was obscured), for whatever good that might do.

[[ Most people who lay claim to public domain works will likely be
discovered by scholars if they ever benefit to any extent, and as a
result such people will likely lose their reputations and be ridiculed
forever more -- none of their works will ever be respected in the
future, even if they were original.  If only the software world could
benefit from the same scholarly pressure the literary world sometimes
does!  BTW, scholars often find the true source for plagiarized works
because the copy contains mistakes only in the source.  There are
parallels in the software world (behaviour at boundary and out-of-bounds
conditions, etc.), and there's also the possibility of using digital
signatures. ]]

> I say this as it was brought to my attention that a number of legal
> journals (reporting public court decisions, etc) copyright the page
> numbers in their documents. They can't copyright the decision (it's
> public), but they can do something to protect the effort they took to
> making the book.

Yes, I'm also very well aware of such things.

What would be important here would be to check and see if there's any
case law against someone trying to violate such a copyright which would
give some legal interpretation to this issue.  That's lawyers for you.
I'd bet that if you started both sides of such a legal case on level
ground that it would *never* end.  I don't personally know of any such
case law though.  These things would be extremely hilarious if it wasn't
for the fact that there are a lot of serious lawyers out there who would
love to profit from such silly things.

Anyway, I believe the "fair use" rules will allow you to distribute
physical copies of *one* public domain article from such a journal,
esp. if you were to distribute a limited number for free, or possibly
even for the cost of duplication and distribution.  I think part of
"fair use" is based on cost of prosecution -- the law has just made it
explicit that trivial violations are not to be worried about.

Of course again if you just re-type the words of the public domain
article then you could even claim to have written it yourself, for
whatever good that would do.  There would be no indication of which
journal you happened to find it in.  That's what "public domain" is all
about.

Mind you, most such journals are not re-printing true public domain
articles, but rather articles that are covered by copyright licenses
which in some cases are very similar to the BSD copyright license.
Indeed the BSD copyright license is much the same one Universities and
Governments use to cover all sorts of works that they wish to be widely
available (presumably to disseminate their content to the public in the
most effective manner).

What it all boils down to is that RMS and the FSF et al, and those who
do similar things, have found ways to create scare tactics in the form
of potentially unenforceable copyright licenses which either force people
to provide "free" (in the FSF's definition) software, or to pay
licensing fees.  Scare tactics like this are most effective when there's
little or no supporting case law and nobody dares to "go first".

I like the idea of "free" (as RMS and the FSF define "free") software.
I applaud the efforts of RMS and the FSF et al to create "free" software
and I offer them the best of wishes in their goals.  However I don't
want to see them fail because they've created un-enforceable licenses.
Nor do I want to see people afraid of "using" free software (i.e.
including as part of their own software distributions) because they're
scared their own creations will be forced to become "free" too, and
that's precisely what I feel the NetBSD Foundation is doing by avoiding
the replacement of a fundamentally flawed and somewhat broken, but
unrestricted DB 1.x library with a vastly improved, fixed, and "freed"
DB 2.x library; or by avoiding GNU GPL copyright device drivers, etc.

(I obviously don't necessarily prefer all GNU software either just
because it's "free", esp. when it's not the superior version or even
when its the only version available but it's not of high quality or in
some way incompatible with or against the grain of the software it is
intended to replace.  I'm extremely happy to see NetBSD eliminate GNU
tar.  I'd love to see GNU awk eliminated (Mawk or AT&T awk please!).  I
don't like GNU coding styles.  I don't like GNU make.  I don't like GNU
command-line options.  There are probably other things I don't like
about GNU software too.)

Unless someone has concrete examples that show how a GNU (L)GPL style
license (or even one like the a Sleepycat version) has actually resulted
in the equivalent of making something "free", I'll leave this as it is.
Past experience in similar debates has proven that those defending the
GPL won't give up preaching to the converted and if there's no case law
to settle these issues then they'll just remain as they are.

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>