Microsoft prohibits GPLed work via licensing of CIFS standards
Posted 5 Apr 2002 at 07:27 UTC by atai
In its continuous battle against the GPL, Microsoft is trying a new
tactic, a combination of patent claims and licensing of technical
standards. In the "Royalty-Free CIFS
Technical Reference License Agreement", Microsoft defines the GNU
GPL as an "IPR Impairing License" and requires companies not to
distribute their implementations of the CIFS specification "in any
manner that would subject such Company Implementation to the terms of
an IPR Impairing License." This attack is clearly aimed at the
successful GPLed CIFS implementation, Samba.
The license defines
1.4 "IPR Impairing License" shall mean the GNU General Public
License, the GNU Lesser/Library General Public License, and any license
that requires in any instance that other software distributed with
software subject to such license (a) be disclosed and distributed in
source code form; (b) be licensed for purposes of making derivative
works; or (c) be redistributable at no charge.
1.6 "Necessary Claims" shall mean those claims of a patent or
patent application, including without limitation, United States Patents
Nos. 5,265,261 and
(a) are owned, controlled or sublicenseable by Microsoft without
payment of a fee to an unaffiliated third party; and (b) are
necessarily infringed by implementing the CIFS communication protocol
as set forth in the Technical Reference, wherein a claim is necessarily
infringed only when there are no technically reasonable alternatives to
And it requires
3.3 IPR Impairing License Restrictions. For reasons, including
without limitation, because (i) Company does not have the right to
sublicense its rights to the Necessary Claims and (ii) Company's
license rights hereunder to Microsoft's intellectual property are
limited in scope, Company shall not distribute any Company
Implementation in any manner that would subject such Company
Implementation to the terms of an IPR Impairing License.
Response?, posted 5 Apr 2002 at 10:31 UTC by tk »
Has the FSF responded to this threat yet?
be swift, posted 5 Apr 2002 at 13:31 UTC by lev »
so "the industry has requested" micro$oft to enforce such
<laugh>. it sounds more appropriate if the first sentence was
written as, "to protect our own interests"...
before we start making fun of this agreement (yes, it's funny to see
micro$oft doing all these shameless things), there are two things that i
think we must do at once.
- have someone with legal expertise to check if it's still legally
continue using past releases of samba. other than that, the
samba people should start planning samba's future. we all don't wish to
this pearl of the open source world, do we?
- every one of us should now check if any of our own projects
implement any non-open standards or protocols. no matter who owns it
they might one day do the same thing. we gotta be prepared, to face
similar problem in the near future. perhaps, we should make it a
practice to get a written agreement before implementing any non-open
standards the open source
here's a few thoughts of mine, after reading the CIFS agreement in
in section 1.2 of the agreement,
1.2 "Company Implementation" shall mean only those portions of the
software developed by Company that implement CIFS for use on
does anyone know why they don't consider open source CIFS
implementations for micro$oft platforms as a threat?
then section 1.4 says, implementations are not to be distributed
in source code. do programs written in interepreted languages
as source codes? i have the feeling that no courts have been confronted
with such a question.
referring to section 3.3, i'm wondering if someone will...
- write a CIFS implementation for a non-micro$soft platform.
- declare it closed source, but leave the source code unguarded on
- then, sell copies of the product for one dollar each.
(OD == open source developer)
M$: why are you violating the agreement?
OD: did i?
M$: yes you did, people got hands on your codes.
OD: oops, i've forgot to set a password for my CVS server!
M$: then why are you selling your products at 1 dollar?
OD: well, i just don't like to make too much money.
that doesn't seem to be violating the agreement. <smile>
PS: tk, it is, in fact, the responsibility of us
I shall be complaining to the UK and European anti-trust authorities
about this, assuming it holds water on closer inspection. I urge
others to perform their own evaluation and make similar complaints.
Re: be swift, posted 5 Apr 2002 at 15:28 UTC by tk »
it is, in fact, the responsibility of us all to respond.
True, but I'd like to know FSF's response (if any) first.
This has got to be the most serious attack Microsoft has pulled off so
longer do they need to protect e.g. their Word document format and wait
someone reverse-engineers it. They can just `publicize' (patent) the
relax for the next 17 years.
The worst that can happen is that there'll be a great chasm between the
Microsoft world and the non-Microsoft world. And the temptation to
non-Microsoft software and enter the Microsoft world will be great,
big organizations (esp. governments) will likely be using Microsoft
people need to communicate with these organizations.
What can be done about this? I don't know (if I do, I'll be a
strategist!). But one
thing is certain: the open-source community must try to break free from
the current situation of constantly chasing Microsoft's tail. Right now,
Microsoft creates SMB, someone tries to clone SMB; when Microsoft
someone tries to clone .NET; when Microsoft creates Word, someone tries
clone Word; and so on. This definitely won't do.
I'll still curious as to FSF's response. :-)
I've probably misinterpreted something, but am I right in
thinking that section 1.4 doesn't hold for the GPL/LGPL?
(which might be why they're explicitly mentioned.)
1.4 "IPR Impairing License" shall mean the GNU General Public License,
the GNU Lesser/Library General Public License, and any license that
requires in any instance that other software distributed with software
subject to such license (a) be disclosed and distributed in source code
form; (b) be licensed for purposes of making derivative works; or (c) be
redistributable at no charge.
The GPL makes no restriction on the license of other software
distributed with GPL'd code. Just because GCC is GPL'd doesn't mean that
Perl has to be GPL'd if you ship the two programs on one CD. Pretty much
every GNU/Linux distro ships these two together, but Perl isn't
"infected" by the GPL.
The end of clause 2 of the GPL states:
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
I'm not really familiar with other OS/Free licenses, but I don't know of
any that impose restrictions on other software distributed with such
licensed software. Wouldn't that mean an FTP server that held such
software would be unable to serve anything
without it coming
under the license?
I can't believe M$ have misunderstood the GPL, so is this just
misleading FUD designed to imply that the GPL does "infect" other
software shipped with it? And if so, isn't using a legal contract as
propaganda a little silly?
, you've completely misinterpreted the document
on MS's website. They define the set of licences they don't like
as the "GPL, the LGPL, and any other licence meeting condition FOO".
That the GPL doesn't meet licence FOO is irrelevant - it's still in
their set of banned licences.
Change the name, posted 5 Apr 2002 at 17:51 UTC by mbrubeck »
As one Slashdot poster suggested,
it is possible to create a free license compatible with the GPL or LGPL that
does not fall under Microsoft's definition of "IPR Impairing." Then as
long as it is not called the GNU General Public License or the GNU
Lesser General Public License, it is legal to use this license for a
CIFS implementation. Two-way compatibility would allow projects
under this license to use existing code from GPL projects and vice-versa.
Therefore the Microsoft licensing terms do nothing to prevent free
software implementations of CIFS. They are pure propaganda, another
attempt by Microsoft to make false implications about the GPL. This is
a PR battle, not a legal or technical one.
This doesn't seem to work, because the license names GPL/LGPL
and anything which is like the GPL. Any license which requires
distribution in source code, access for modification (derivative works),
or require redistribution at no cost.
I think the BSD license would work, but not a renamed GPL.
I don't think it's so much as PR, as allowing MS to do what they want
with their patents. Say GPL was allowed, but Microsoft makes
proprietary software. They give you access to patents and you develop
software based on their patents under GPL. GPL won't let them use the
software (as they like: proprietary) you developed and integrate that w/
whatever they wish. It's "you scratch my back, I'll scratch your's." I
think they are giving you the ability to develop free software
(BSD-style freedom), and in return you grant Microsoft the ability to
reincorporate the software into whatever they wish (this freedom comes
from the BSD-style license).
The GPL removes a freedom that Microsoft depends on: the ability to
create proprietary software. I don't see any other way Microsoft can be
"open source friendly" without being anti-GPL at the same time. They
are too dependent on proprietary software to get cozy with the GPL.
Aside from the (L)GPL, the Microsoft agreement forbids [my emphasis]:
any license that requires in any instance that other software
distributed with software subject to such license (a) be disclosed
and distributed in source code form; (b) be licensed for purposes of
making derivative works; or (c) be redistributable at no charge.
The GNU GPL places no restrictions on other software distributed with
GPL-covered software. Separate works distributed together are
specifically exempted from section two of the GPL:
If identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. [...]
In addition, mere aggregation of another work not based on the
Program with the Program (or with a work based on the Program) on a
volume of a storage or distribution medium does not bring the other work
under the scope of this License.
Even if "other software distributed with" is taken to include the
derivative works covered by the GPL,
there is still the LGPL. Code released with this license can be
distributed as part of a single binary with non-LGPL or even non-free
code. In no way do any of the three named restrictions apply
to "other software distributed with software subject to" the LGPL. A
license whose restrictions are equivalent to those in the LGPL would not
be forbidden by the Microsoft agreement as written.
re: be swift, posted 6 Apr 2002 at 04:12 UTC by julian »
, the second point of action you note is a very
important one that I feel nowhere near enough people in the Open Source
movement take seriously. I realize I'm heavily biased, but those in the
Open Source community who use AIM, ICQ, Yahoo!, or MSN IM really need to
hear what you're saying. The companies which run these protocols could
at any time decide to make it illegal for people to make Open Source
I don't get it, posted 6 Apr 2002 at 07:15 UTC by ali »
I don't understand the non-GPL-but-IPRImpairing-part.
What license fulfills any of (a) to (c)? Doesn't it mean a (hypothetic)
license of software X stating "If you put me, X, into the same box with
software Y and distribute it, Y must have such a license"
Did I understand that correct? Does such a license exist for
re: be swift, posted 6 Apr 2002 at 07:49 UTC by tk »
I realize I'm heavily
biased, but those in the Open Source community who use AIM, ICQ, Yahoo!,
or MSN IM
really need to hear what you're saying.
Many of these people use these protocols simply because they need to
communicate with the Microsoft world... It's a tough nut to crack.
I've said it so many times before, but it is worth repeating: Shared
Source is only a means by which to contaminate as many Open Source
projects as possible. Even if it's just a tiny amount, say, one
developer saw Microsoft code at his company five years ago, it's enough
to bring the wrath of Redmond down on your head. That's why the Samba
project has been so paranoid about their contributors.
I think lev is on the mark with his point #2. Maybe it's time to create
an authentication-based network file system protocol akin to CIFS, but
completely Open Source. (By akin, I mean it looks and feels
the UI level, but the inner working is completely different.) And it
should work on both Windows and *nix. Of course, maybe there is already
such a creature; if someone knows of one, I'm willing to be enlightened.
You put a technical reference somewhere, and you say "You can't
implement it if you're going to do the public a favor".
I could write another large article and tell why this is bad. And we
could all make sense of it. Who gives the right to MS for *making* law
via a simple license? That's a severe flaw in our law system.
So instead I just want to say that this is fucking bullshit. Fuck that
license and the deepshit who thought of it. Programmers at Microsoft
should be ashamed of their imbecile managers.
re: be swift, posted 6 Apr 2002 at 14:12 UTC by lev »
julian, well... as tk has written,
"it's a tough nut to crack". tho' we never meant to be fully compatible
with those proprietary standards, we do want to be at least able to
communicate. in this case, the developers are having a hard time.
personally, i pay extra respects to those developers who implement
proprietary standards on open source platforms. their works really worth
an (if not more) honour. imagine if we're to live without them. yes, we
can still live, but it'll be rather painful.
exa, there are, of course, some absurdities in our
current law system. the patent laws is in such a way that favour the
patent owners. the reason _was_ "to encourage innovations in the
fields of science and technology". but in recent years, the laws
to be going against the interests of humanity as a whole.
to make changes in the patent laws is really beyond my reach. i can
only hope that it will soon have a change for the better. but even with
the current patent laws in place, i still feel that one shouldn't make
such a move as M$ did. yes, they have such 'right' under the
laws, but it is no doubt _unethical_.
prior art, posted 7 Apr 2002 at 05:05 UTC by lev »
wrote in an email to me,
In response to, "have someone with legal expertise to check if it's
still legally safe to continue using past releases of samba.", see
"prior art". I'll explain it - but I not 100% sure of it. Basically it
has to do
with things that were made befor, using patented things, befor the
patent was issued etc, or things done independantly.
I'd recommend talking to a lawyer, most would proboly give a nice
explination of it all for free, considering it wouldnt take alot of
there time - and they might assume that you are querying about a topic
which might lead to you
being a client from which they will have some sorta court case or
otherwise from that simple query, worth a try =).
Well, no wonder MS are so keen to have the EPO accept software patents.
I think this case might be the one that we can use to encourage
mainstream businesses in Europe to oppose software patents.
A lot of businesses, financial institutions etc. use samba to glue their
networks together, so we can point out that MS are trying to stop this
in the USA, and will do similar evils in Europe given the chance.
It makes it much easier to demonstrate that Software Patents are most
useful as a tool to the benefit of illegal monoplists, and damage the
efficiency normal business.
It's also good that it's not a pure "Free Software" issue. Many people
use samba to link their high end commercial Unix servers into their
networks, so the normal tactic of the patent lawyers to dismiss us as
ranting Free Software fanatics, will (hopefully) not work here.
Thanks atai. However, personally I'm disappointed by
response. After so many years, the FSF should've known better than to
harping on "software freedom".
to create a nice little conflict within microsoft's clause:
1) publish the intellectual property with a license agreement attached
saying "before you read this document outlining our intellectual
must sell us your soul".
this is exactly what microsoft attempts to do with _their_ intellectual
property. not only is it a dig at their stupid attitude which would, if
they challenged it in court, allow us to challenge THEIR use of the same
techniques, but also it appears to be a "protection of intellectual
2) publish the source code under the GPL or any other free software
license, or a renamed GPL, such that it conflicts with the principles of
the clause 1.4.
in two easy steps, you have fucked them over in knots of their own
the second thing is - the GPL is _not_ a non-chargeable license. you
_may_ charge for distribution of the media.
Not enough, posted 16 Apr 2002 at 16:11 UTC by tk »
The keyword here is "patent
". "Intellectual property" isn't
enough. Even if
you publish the details of some protocol ABC but require others to sell
in order to read it, people are still allowed to reverse-engineer ABC
and come up
with a clean-room implementation, and there's nothing you can do. With
it's different: even clean-room implementations aren't allowed.
patents, ip, posted 17 Apr 2002 at 09:08 UTC by lkcl »