Subject: A statement on NPL/MPL/GPL licensing issues
Organization: Columbia University Law School
At the request of Richard Stallman I am writing to express legal views
on several questions recently raised about the relations among the
Free Software Foundation's General Public License (GPL), the Netscape
Public License (NPL) version 0.95 and the Mozilla Public License (MPL)
0.95. I need to emphasize that although I act as Outside General
Counsel to the Free Software Foundation, in the comments that follow I
am expressing my personal opinion as to the licenses themselves and
their legal consequences.
In this posting I address three questions:
1. Can the NPL be made to coexist with GPL-covered code by minor
modifications of the NPL not in any way harmful to Netscape's
2. Could Netscape release the Netscape Communicator under the GPL?
3. Could Netscape release the Netscape Communicator under both the
GPL and the NPL simultaneously?
Questions 2 and 3 have been much discussed since the original release
of the draft NPL, and I think it will be helpful to explain the legal
position of those of us who administer the GPL on behalf of FSF. But
although I think that some of Netscape's statements about the
possibility of GPL release overstate the supposed difficulties of
using the GPL for such a project, the first issue, which is quite a
different one, seems to me the most important.
The problem that concerns me, and others interested in encouraging the
broadest possible free software development, is that there is a great
deal of very useful code distributed under the GPL, and many
developers will want to construct Larger Works containing combinations
of NPL-covered code distributed by Netscape and others with
GPL-covered code. Version 0.95 of the NPL makes it more difficult
than necessary for free software authors to distribute works
containing both NPL-covered and GPL-covered code.
FSF seeks two amendments to the NPL that would, in my opinion, resolve
the problem in a way that is completely non-injurious to Netscape's
interests. The two paragraphs would be added to the NPL following the
current Larger Works provision 3.7:
3.7.1. You may distribute a Covered Work under the terms of the GNU
General Public License, version 2 or newer, as published by the Free
Software Foundation, when it is included in a Larger Work which is
as a whole distributed under the terms of the same version of the
GNU General Public License.
3.7.2. If you have received a copy of a Larger Work under the terms of a
version or a choice of versions of the GNU General Public License,
and you make modifications to some NPL-covered portions of this
Larger Work, you have the option of altering these portions to say
that their distribution terms are that version or that choice of
versions of GNU General Public License.
Under these provisions, developers could combine NPL-covered code with
GPL-covered code to create Larger Works for distribution under the
GPL. Such Larger Works would not be available for proprietary
redistribution by Netscape, which is less advantageous to Netscape
than distribution under the NPL. But the position of such Larger
Works is no different in this respect than if the Works were
distributed under the MPL, which also denies Netscape preferential
rights. As a result, Netscape would lose nothing by these amendments
not already offered under the MPL, while a developer wishing to make a
larger work combining NPL-covered code with such GPL-covered code as
GNU Emacs or Samba would be able to do so. I believe that this result
will bring about the most productive possible integration of Netscape
Communicator into the free software community. I hope that Netscape's
counsel will agree that this change is superior for all parties, and
will agree to modify the NPL accordingly.
This question is of immediate practical significance. The other two
questions, whether Netscape could release the Communicator under the
GPL instead of the NPL, or under both licenses simultaneously, seem to
me academic in nature. Netscape has chosen not to take either course,
which is a decision FSF understands with regret. But because some of
the statements made by Netscape or its employees to explain that
decision seem to me to exaggerate the difficulties of using the GPL, I
want to clarify the legal situation involving the use of the license.
Both the Netscape-distributed FAQ and comments by Mr. Mark Shaver have
suggested that the presence of cryptographic code in the Netscape
Communicator somehow prevented distribution of the code under the
GPL. This is not correct. Cryptographic software, including that
incorporated in Netscape Communicator, is subject to export controls
implemented by the United States Government under the Arms Export
Control Act. The regulations prohibit unlicensed exportation of
cryptographic software. They do not prohibit distribution within the
United States and Canada of source code for cryptographic software.
They do not impose any restraints on license terms for software
distributed within the United States and Canada, nor are there any
license terms that can render software covered by the regulations
freely exportable without license.
The GPL has been used to distribute non-exportable cryptographic
software. Versions of PGP, the strongest general-use cryptographic
software freely available in the United States, have been distributed
by MIT and others under the GPL. I represented Philip Zimmerman, the
author of PGP, in connection with the United States Government's
investigation of PGP, and I participated in the discussion concerning
licensed distribution of PGP by MIT. I can state as a formal legal
opinion that use of the GPL for distribution of cryptographic software
in no way changes the developer's rights and duties under US export
control law. Whether Netscape releases its code under GPL or NPL, the
situation is precisely the same.
Mr Shaver and the Netscape FAQ have also referred to Netscape's
contractual obligations to distribute source code to certain
contractual partners on non-free terms. The statements imply that if
Netscape were to release the Communicator code on non-free terms to
some parties, as it is contractually obliged to do, it could not
release that code under the GPL to others.
This statement also is incorrect. Anyone receiving source code under
the GPL must redistribute that code under GPL terms. But Netscape
does not receive its own code under the GPL. As the author of the
relevant work, it may release its code both on non-free terms to some
and under the GPL to everyone else. Preexisting contractual
obligations to provide source code on non-GPL terms do not preclude
simultaneous release of code under the GPL, as FSF has previously
advised other developers under similar circumstances.
It has also been stated that Netscape does not own sufficient rights
to some of the code included in the Communicator to release that code
under the GPL. Because neither the Netscape FAQ nor Mr Shaver have
provided particulars of the licenses involved, I cannot state any
opinion on this point. But I am not able to ascertain which rights
Netscape would need to have in order to release particular third-party
code under the GPL that are not called for in order to release
the same code under the NPL. I wish that at some point in the
conversation Netscape had specified at least one such example in order
to make clear the basis of the restriction.
Eben Moglen voice: 212-854-8382
Professor of Law & Legal History fax: 212-854-7946 moglen@
Columbia Law School, 435 West 116th Street, NYC 10027 columbia.edu
From: robert havoc pennington <h...@pobox.com>
Subject: Re: A statement on NPL/MPL/GPL licensing issues
References: <firstname.lastname@example.org> <351C0974.B17A8898@netscape.com>
Organization: Another Netscape Collabra Server User
Mike Shaver <sha...@netscape.com> writes:
> How is it superior for Netscape, precisely?
> I can see all sorts of benefit for the GNU/GPL community, but the
> concrete benefits for Netscape elude me.
I'll stay out of most of this, :), but I thought I'd suggest an answer
to this question:
If it gets more people to use the code in more ways, you get more bug
reports. If people want to use the code yet have you continue to
maintain it, they will have an incentive to rewrite it in a more
modular way, helping Mozilla as a whole. Finally, it is good PR in a
certain community, though perhaps bad PR in another; only Netscape can
decide how to balance that.