RE: IETF Trust FAQ

At Harald's suggestion that others "speak for
themselves," I'll speak up. I
agree with Simon. I'd feel more comfortable about this
working group punting
Simon's issues and passing them to the IETF Trust if I had
some assurance
that they were going to solve these IP policy problems in a
public way with
public input. 

I do not believe that the current draft of RFC 3978 solves
the problems
Simon and others have identified. For the record, I'm
copying an earlier
private thread with Simon. I copied Brian and Jorge but they
haven't
responded. So I'm going to copy it to the full list below.
We might as well
discuss these problems now rather than have to do it right a
second time.

I wish the IETF Trust lawyers who promise to help that
organization draft
its policies were here now to help us, because our current
policies aren't
sufficient to do the job.

/Larry Rosen

*********************************

Thanks, Simon. I specifically wrote to you and Brian
directly so that I can
get to the gist of the problem you wrote about without
in-channel "noise."
 I
look forward to Brian's response. What I ultimately say
publicly on
the list depends upon what I learn from you two.

I agree, though, with the fundamental premise you raised in
your earlier
email, and so I suggested that contributors to IETF should
be given express
notice and assurance that their contributions are intended
to be
*sublicensed to the public by the IETF Trust* in the form of
published
specifications compatible with both open source and
proprietary licenses.

> Stating your opinion of whether the RFC 3978 license
gives third
> parties the right to extract and use code portions from
RFCs to the
> IPR WG list would be useful.  I think it is problematic
from several
> views, but Brian and the IETF Trust appear to be
pushing for an
> interpretation that says "there is no problem
here, go away".

RFC 3978 states that "[the] rights addressed in this
document fall into the
following categories: rights to make use of contributed
material...." 

The words "make" and "use" come from
patent law, not copyright law.
Copyright addresses the relatively trivial problem of
copying expressive
code into a computer's memory but it doesn't affect what
happens when you
press start. For that you may need patent rights.

We can't wish that demon to go away by saying simply, as we
currently do in
RFC 3978, section 3.3(a)(E), that the licenses "granted
under this paragraph
(E) shall not be deemed to grant any right under any
patent...." Every
modern open source and proprietary license contains a clear
statement of
patent rights being granted along with distribution of the
copyrighted code
so that licensees know what they may make or use.

This problematic statement in 3.3(a)(E) is apparently
contradicted by the
last words of the leading paragraph 3.3(a), which refers to
"all
intellectual property rights in the Contribution."
Certainly that includes
patents?

I'm confused by the reluctance to address patents head-on in
this document.

/Larry

P.S. Since I am stating my views of legal language and
effect, I will now
copy Jorge.

> -----Original Message-----
> From: Simon Josefsson [mailto:simonjosefsson.org]
> Sent: Monday, January 15, 2007 1:47 PM
> To: lrosenrosenlaw.com
> Cc: 'Brian E Carpenter'
> Subject: Re: IETF Trust FAQ
> 
<snip>
> 
> See RFC 3978 section
> 3.3 of <http://www.i
etf.org/rfc/rfc3978.txt>. 
>  
> If I understand correctly, the current goal is not to
write legal
> text, but to document the IETF's intention, and let the
IETF Trust
> craft the legal text, with help from lawyers.  I don't
expect the
> IETF community to be able to influence or review the
end result.  I'd
> love to be wrong on that, or see the process change to
something more
> open and something that have any chance of resulting in
a workable
> outcome, but there seem to be no interest in exploring
any options.
> 
> Stating your opinion of whether the RFC 3978 license
gives third
> parties the right to extract and use code portions from
RFCs to the
> IPR WG list would be useful.  I think it is problematic
from several
> views, but Brian and the IETF Trust appear to be
pushing for an
> interpretation that says "there is no problem
here, go away".
> 
> /Simon


_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg

Jorge Contreras wrote:
> However, if a license is granted
> under patents under 3.3.a, it is very limited and only
extends to
> standards
> development work within the IETF.

That is precisely my point, and perhaps Simon's also. Well
said!

> The licenses under 3.3.a (other than 3.3.a.E, see
below)
> do NOT grant any rights to implement IETF standards
> in products.  IETF Contributors are NOT required to
grant patent
> licenses to implementers.  This has always been IETF's
policy, and
> there is no ambiguity here.

And that is why this is, and always has been, a policy
issue. You can't
solve this by merely restating the previous position of IETF
that you don't
care about receiving patent grants from contributors for the
making and
using of their contributions in both open source and
proprietary
implementations. This isn't just an engineering game. We
create standards to
enable functional implementations, not for the expressive
beauty of the
specifications.

> Finally, please note that these policies were formed
prior
> to the establishment of the IETF Trust.  I don't
understand why
> the Trust is being criticized here, or why the Trust is
expected
> to do something different that what's stated in 3978.

No ad hominem criticism intended. Again, this is an
important *policy*
issue.

> Of course, the community is free to change the rules as
it desires.
> If a change to the above structure is desired by the
community,
> such a change would not be difficult to implement. 
However, I think
> it is disingenuous to claim that ambiguities exist in
3978 and
> that these ambiguities support an argument for change.

No, that's not the argument for change. We need change
because the current
IETF policy does not guarantee that every contributor's
necessary IP is
licensed to the public for implementation of IETF
specifications in both
proprietary and open source software. 

Without that they are not open standards. Without that they
are just pretty
documents available under IETF copyright licenses. Thanks
but that's not
enough!

/Larry Rosen


> -----Original Message-----
> From: Contreras, Jorge [mailto:Jorge.Contreraswilmerhale.com]
> Sent: Tuesday, January 16, 2007 5:35 PM
> To: lrosenrosenlaw.com; ipr-wgietf.org
> Subject: RE: IETF Trust FAQ
> 
> 
> Larry,
> 
> I think that you are attempting to find ambiguity in
the language
> of RFC 3978 when there is none.
> 
> Section 3.3.a deals with the licenses that
Contributors
> grant to IETF participants with respect to their
Contributions
> (which are mostly text, but can also be code).  These
> licenses enable the standards development and
publication work
> that is coordinated through IETF.  To the extent that a
patent
> license may be required to conduct IETF standards
development
> work (a proposition that is debatable), such a license
is granted.
> That's why 3.3.a says "under all intellectual
property
> rights in the Contribution".  However, if a
license is granted
> under patents under 3.3.a, it is very limited and only
extends to
> standards
> development work within the IETF.
> 
> The licenses under 3.3.a (other than 3.3.a.E, see
below)
> do NOT grant any rights to implement IETF standards
> in products.  IETF Contributors are NOT required to
grant patent
> licenses to implementers.  This has always been IETF's
policy, and
> there is no ambiguity here.
> 
> Section 3.3.a.E covers a special case in which code is
embedded
> in IETF documents.  Under this license, code can be
extracted
> and used.  However, there is an explicit exception in
the license:
> it does not include a license under patents.  That
exception was
> included for very reason stated above.  IETF does not
require
> participants to grant patent licenses to implementers. 
Someone who
> extracts code from an RFC for use in a product is an
implementer,
> and that implementer does not automatically get a
license under
> the patent.  Arguing that the implementer gets a patent
license
> is contrary to both the express wording of 3.3.a.E, as
well as
> IETF's general policy relating to patent licenses.
> 
> Finally, please note that these policies were formed
prior
> to the establishment of the IETF Trust.  I don't
understand why
> the Trust is being criticized here, or why the Trust is
expected
> to do something different that what's stated in 3978.
> 
> Of course, the community is free to change the rules as
it desires.
> If a change to the above structure is desired by the
community,
> such a change would not be difficult to implement. 
However, I think
> it is disingenuous to claim that ambiguities exist in
3978 and
> that these ambiguities support an argument for change.
> 
> Jorge
> 
> 
> 
> > -----Original Message-----
> > From: Lawrence Rosen [mailto:lrosenrosenlaw.com]
> > Sent: Tuesday, January 16, 2007 7:40 PM
> > To: ipr-wgietf.org
> > Subject: RE: IETF Trust FAQ
> >
> >
> > At Harald's suggestion that others "speak for
themselves,"
> > I'll speak up. I
> > agree with Simon. I'd feel more comfortable about
this
> > working group punting
> > Simon's issues and passing them to the IETF Trust
if I had
> > some assurance
> > that they were going to solve these IP policy
problems in a
> > public way with
> > public input.
> >
> > I do not believe that the current draft of RFC
3978 solves
> > the problems
> > Simon and others have identified. For the record,
I'm copying
> > an earlier
> > private thread with Simon. I copied Brian and
Jorge but they haven't
> > responded. So I'm going to copy it to the full
list below. We
> > might as well
> > discuss these problems now rather than have to do
it right a
> > second time.
> >
> > I wish the IETF Trust lawyers who promise to help
that
> > organization draft
> > its policies were here now to help us, because our
current
> > policies aren't
> > sufficient to do the job.
> >
> > /Larry Rosen
> >
> > *********************************
> >
> > Thanks, Simon. I specifically wrote to you and
Brian directly
> > so that I can
> > get to the gist of the problem you wrote about
without
> > in-channel "noise."
> >  I look
forward to Brian's response. What I ultimately say
> > publicly on
> > the list depends upon what I learn from you two.
> >
> > I agree, though, with the fundamental premise you
raised in
> > your earlier
> > email, and so I suggested that contributors to
IETF should be
> > given express
> > notice and assurance that their contributions are
intended to be
> > *sublicensed to the public by the IETF Trust* in
the form of published
> > specifications compatible with both open source
and
> > proprietary licenses.
> >
> > > Stating your opinion of whether the RFC 3978
license gives third
> > > parties the right to extract and use code
portions from RFCs to the
> > > IPR WG list would be useful.  I think it is
problematic from several
> > > views, but Brian and the IETF Trust appear to
be pushing for an
> > > interpretation that says "there is no
problem here, go away".
> >
> > RFC 3978 states that "[the] rights addressed
in this document
> > fall into the
> > following categories: rights to make use of
contributed material...."
> >
> > The words "make" and "use"
come from patent law, not copyright law.
> > Copyright addresses the relatively trivial problem
of copying
> > expressive
> > code into a computer's memory but it doesn't
affect what
> > happens when you
> > press start. For that you may need patent rights.
> >
> > We can't wish that demon to go away by saying
simply, as we
> > currently do in
> > RFC 3978, section 3.3(a)(E), that the licenses
"granted under
> > this paragraph
> > (E) shall not be deemed to grant any right under
any patent...." Every
> > modern open source and proprietary license
contains a clear
> > statement of
> > patent rights being granted along with
distribution of the
> > copyrighted code
> > so that licensees know what they may make or use.
> >
> > This problematic statement in 3.3(a)(E) is
apparently
> > contradicted by the
> > last words of the leading paragraph 3.3(a), which
refers to "all
> > intellectual property rights in the
Contribution." Certainly
> > that includes
> > patents?
> >
> > I'm confused by the reluctance to address patents
head-on in
> > this document.
> >
> > /Larry
> >
> > P.S. Since I am stating my views of legal language
and
> > effect, I will now
> > copy Jorge.
> >
> > > -----Original Message-----
> > > From: Simon Josefsson [mailto:simonjosefsson.org]
> > > Sent: Monday, January 15, 2007 1:47 PM
> > > To: lrosenrosenlaw.com
> > > Cc: 'Brian E Carpenter'
> > > Subject: Re: IETF Trust FAQ
> > >
> > <snip>
> > >
> > > See RFC 3978 section
> > > 3.3 of <http://www.i
etf.org/rfc/rfc3978.txt>.
> > >
> > > If I understand correctly, the current goal
is not to write legal
> > > text, but to document the IETF's intention,
and let the IETF Trust
> > > craft the legal text, with help from lawyers.
 I don't expect the
> > > IETF community to be able to influence or
review the end
> > result.  I'd
> > > love to be wrong on that, or see the process
change to
> > something more
> > > open and something that have any chance of
resulting in a workable
> > > outcome, but there seem to be no interest in
exploring any options.
> > >
> > > Stating your opinion of whether the RFC 3978
license gives third
> > > parties the right to extract and use code
portions from RFCs to the
> > > IPR WG list would be useful.  I think it is
problematic from several
> > > views, but Brian and the IETF Trust appear to
be pushing for an
> > > interpretation that says "there is no
problem here, go away".
> > >
> > > /Simon
> >
> >
> > _______________________________________________
> > Ipr-wg mailing list
> > Ipr-wgietf.org
> > https:/
/www1.ietf.org/mailman/listinfo/ipr-wg
> >
> 
> _______________________________________________
> Ipr-wg mailing list
> Ipr-wgietf.org
> https:/
/www1.ietf.org/mailman/listinfo/ipr-wg


_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg

On Tue, 16 Jan 2007 18:39:36 -0800
"Lawrence Rosen" <lrosenrosenlaw.com> wrote:

> Jorge Contreras wrote:
> > However, if a license is granted
> > under patents under 3.3.a, it is very limited and
only extends to
> > standards
> > development work within the IETF.
> 
> That is precisely my point, and perhaps Simon's also.
Well said!
> 
> > The licenses under 3.3.a (other than 3.3.a.E, see
below)
> > do NOT grant any rights to implement IETF
standards
> > in products.  IETF Contributors are NOT required
to grant patent
> > licenses to implementers.  This has always been
IETF's policy, and
> > there is no ambiguity here.
> 
> And that is why this is, and always has been, a policy
issue. You
> can't solve this by merely restating the previous
position of IETF
> that you don't care about receiving patent grants from
contributors
> for the making and using of their contributions in both
open source
> and proprietary implementations. This isn't just an
engineering game.
> We create standards to enable functional
implementations, not for the
> expressive beauty of the specifications.
> 
Larry, are you suggesting that we reopen the patent issue
within this
working group?  If so, I'm afraid that my answer, as chair,
is "no".
We're not chartered to consider the question.  When we
discussed it at
some length in early 2003, the very strong consensus of the
group was
that the IETF should not change its patent policy.  I do not
know of
any major changes in the law or the technical culture around
us that
would lead to different outcome.  I'm also unconvinced that
enough time
has passed that it makes sense to reopen the question yet.

We do have several tasks we're supposed to complete.  I hope
we'll soon
see the documents we discussed in San Diego; at that point,
perhaps we
can wrap up this WG -- it's gone on far too long.

I'm certainly not saying that the IETF's patent policy can
never
change.  I am saying that it would need at least a new
charter, and
possibly a different working group (and almost certainly at
least one
new chair, but that's another matter....).  For either, the
usual rules
would apply: a demonstration to the General Area AD and the
IESG as a
whole that there was sufficient community interest to make
it worth
everyone's time.  Speaking personally and not as chair, I do
not think
there is such interest at this time.  I am, though, quite
certain that
it's not something that this group could undertake at this
time.

So -- until we've completed our chartered tasks, I'm ruling
discussion
(on this mailing list) of the basic IETF patent policy out
of scope.
When we're done, if there's sufficient interest we can
revisit that
question -- but not now.


		--Steve Bellovin, http://www.cs.columbi
a.edu/~smb

_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg

--On Wednesday, 17 January, 2007 12:49 -0800 Lawrence Rosen
<lrosenrosenlaw.com> wrote:

> Ted Hardie wrote:
>> That is why it is key that the IETF remain
committed to open
>> standards in the sense that has been true here all
along:
>> open participation and open availability.  That
makes sure
>> anyone who cares can have a voice. Efforts to
pre-dispose the
>> playing field toward specific patent/license
regimes *reduce*
>> the openness of the IETF, as they discourage some
players who
>> would otherwise participate.
> 
> Unfortunately that way you define "open
availability" doesn't
> ensure that IETF standards can be implemented in open
source
> software. Those standards aren't open enough, AFAICare.

> 
> But my earlier point was broader than that even: This
proposed
> IETF policy doesn't even work for private companies
seeking
> proprietary solutions. There is no advantage to
participate in
> setting industry standards that come with no promise
even of
> RAND patent licenses, much less free ones. 

Certainly individuals sensitive to the needs and concerns
of
such a company should vigorously oppose any WG action that
would
develop a standard whose provisions require practicing a
patent
for which there are no assurances that licenses will be
available.

> If the owners of private patents want to play with
other
> companies in private arenas for private standards,
that's fine
> with me. But that is not what I see as IETF's mission,
or
> rather it is not part of any mission I consider
important for
> free and open source software. 
> 
> Leaving these matters, as you recommend, to the whims
and
> wishes of individual IETF working groups and later
licenses
> TBD isn't a solution for anyone. Decide now that the
IETF is a
> form of public trust and not just an engineer's
playground*,
> and act accordingly.

Assuming that the decisions of those WGs are
"whims" is either
more hyperbole or deeply insulting to the people who
participate
in them.  What I've seen in the IETF, in practice, is a
significant distaste for encumbered technologies coupled
with a
willingness to face reality when something sufficiently
important cannot be standardized without dependence on
something
encumbered. The Public Key Crypto issue mentioned earlier
stands
out as an example and I believe that we made the right
decision,
rather than ignoring the issues and waiting for all of the
relevant patents to expire (which appeared to be the only
other
choice).  You may disagree, of course.

> * Forgive me, my engineer friends, for this bit of
hyperbole. 

And forgive us, including those of us who are not engineers,
for
resisting having the IETF turn into a playground for lawyers
who
want us to adopt sweeping policy rules that would prevent
us
from making finely-tuned decisions in the best interests of
the
Internet.

       john


_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg

Isnt the real issue the precedents set to date. I think the
bigger question 
is whether the IETF can force new standards process
participants to fork 
over their rights to the Trust when the IETF didn't do that
previously.

Further, what happens to IP relative to BCP 78 and 79 when a
revised RFC is 
submitted or a standard which was issued under the
previously in place 
RFC2026 guidelines. I think the current terms for
participation are pretty 
egregious in comparison and forcing any new submissions as
updates to 
previously published works may have issues. IP Extortion is
a term that 
comes to mind pretty easily here.

Todd Glassey

----- Original Message ----- 
From: "Lawrence Rosen" <lrosenrosenlaw.com>
To: "'Ted Hardie'" <hardiequalcomm.com>; <ipr-wgietf.org>
Sent: Wednesday, January 17, 2007 12:49 PM
Subject: RE: IETF Patent policy (was RE: IETF Trust FAQ)


> Ted Hardie wrote:
>> That is why it is key that the IETF remain
committed to open standards
>> in the sense that has been true here all along: 
open participation and
>> open availability.  That makes sure anyone who
cares can have a voice.
>> Efforts to pre-dispose the playing field toward
specific patent/license
>> regimes *reduce* the openness of the IETF, as they
discourage some
>> players who would otherwise participate.
>
> Unfortunately that way you define "open
availability" doesn't ensure that
> IETF standards can be implemented in open source
software. Those standards
> aren't open enough, AFAICare.
>
> But my earlier point was broader than that even: This
proposed IETF policy
> doesn't even work for private companies seeking
proprietary solutions. 
> There
> is no advantage to participate in setting industry
standards that come 
> with
> no promise even of RAND patent licenses, much less free
ones.
>
> If the owners of private patents want to play with
other companies in
> private arenas for private standards, that's fine with
me. But that is not
> what I see as IETF's mission, or rather it is not part
of any mission I
> consider important for free and open source software.
>
> Leaving these matters, as you recommend, to the whims
and wishes of
> individual IETF working groups and later licenses TBD
isn't a solution for
> anyone. Decide now that the IETF is a form of public
trust and not just an
> engineer's playground*, and act accordingly.
>
> /Larry Rosen
>
> * Forgive me, my engineer friends, for this bit of
hyperbole.
>
>
>> -----Original Message-----
>> From: Ted Hardie [mailto:hardiequalcomm.com]
>> Sent: Wednesday, January 17, 2007 11:09 AM
>> To: lrosenrosenlaw.com; ipr-wgietf.org
>> Subject: IETF Patent policy (was RE: IETF Trust
FAQ)
>>
>> >At 8:39 PM -0800 1/16/07, Lawrence Rosen
wrote:
>> >Then I suppose we should take this decision
public and see whether that
>> >encourages better results. I am absolutely
opposed to any IETF IP policy
>> >that results in industry standards that have
fundamental patent
>> impediments
>> >to their implementation in open source--and for
that matter in
>> >proprietary--software. Jorge just confirmed
this to the group, in case
>> >anyone doubts that our current policy has this
effect.
>>
>> >I'd rather see this WG dissolve right now than
to see it publish an IP
>> >policy that doesn't work for anyone.
>>
>> The patent policy of the IETF boils down to
"The WG decides based on
>> information
>> which its participants are required to
provide".  WGs can and do decide
>> to decline to make standard technologies that have
fundamental patent
>> impediments.  They also can, and have, decided that
a particular
>> technology is so useful that it should be made
standard despite
>> the fundamental patents--public key crypto being
the most cited
>> example.  The key aspect of the IETF's policy is
that it is the
>> participants
>> who make the judgement call.  If someone invents a
faster-than-light
>> data transport based on quantum entanglement,
patents it, then asks for
>> the IETF's help in making IP run on it, the IETF
participants interested
>> get to decide whether having FTL transport is worth
dealing with the
>> patent.  It isn't pre-decided for them; it's a
judgement call.
>>
>> That is why it is key that the IETF remain
committed to open standards
>> in the sense that has been true here all along: 
open participation and
>> open availability.  That makes sure anyone who
cares can have a voice.
>> Efforts to pre-dispose the playing field toward
specific patent/license
>> regimes *reduce* the openness of the IETF, as they
discourage some
>> players who would otherwise participate.
>>
>> Ted Hardie
>>
>>
>>
>>
>> _______________________________________________
>> Ipr-wg mailing list
>> Ipr-wgietf.org
>> https:/
/www1.ietf.org/mailman/listinfo/ipr-wg
>
>
> _______________________________________________
> Ipr-wg mailing list
> Ipr-wgietf.org
> https:/
/www1.ietf.org/mailman/listinfo/ipr-wg 


_______________________________________________
Ipr-wg mailing list
Ipr-wgietf.org
https:/
/www1.ietf.org/mailman/listinfo/ipr-wg