www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jeffrey Thompson <jt...@us.ibm.com>
Subject Re: Open Source, Patents and "Patent Exhaustion"
Date Fri, 11 Apr 2008 12:57:41 GMT

"Geir Magnusson Jr." <geir@pobox.com> wrote on 04/11/2008 12:05:22 AM:
> > The good news is that "Work" is very broadly defined.  Basically,
> > anything
> > distributed under the Apache License, so the result is that except for
> > highly hypothetical edge cases, the license does exactly what I think
> > Apache intends it to do.  The license basically follows the code and
> > any
> > Work that it ends up in is licensed.  (Don't jump on me, I know I'm
> > skipping steps in the analysis.)
>
> Can you provide the steps for those of us that are slow?  The defn of
> "Work" in AL2 says :
>
>    "Work" shall mean the work of authorship, whether in Source or
>    Object form, made available under the License, as indicated by
>    a copyright notice that is included in or attached to the work
>    (an example is provided in the Appendix below).
>
> I read that definition to mean that the Work is a specific work of
> authorship (the first "the" gives me the hint) and not any work of
> authorship made available under the License.

In your example of Intel submitting a C language routine (lets call it
"ISORT") that implements their patented sorting algorithm, ISORT would be a
Work.  So would the combined project that ISORT is added to.  The term
"Work" by itself is generic and doesn't necessarily correspond to project
boundaries.

>
> Further, the patent grant says that you get a license
>
> "... where such license applies only to those patent claims licensable
>    by such Contributor that are necessarily infringed by their
>    Contribution(s) alone or by combination of their Contribution(s)
>    with the Work to which such Contribution(s) was submitted."
>
> Now, I can find a few nits to pick here with using this for my
> argument, such as noting the intention was that this phrase only
> defines WHICH patent claims are covered, but it's clear that there is
> a Work, it's a single work - not any work -  and I can at least infer
> it's the work of authorship to which the Contribution was submitted
> rather than some other work.

You are right.  This phrase only describes WHICH patent claims are covered.
It doesn't limit the scope of the grant to the covered claims.  That's why
the language "Work to which such Contribution(s) was submitted" is not
replicated earlier in the paragraph.  The scope of the grant is broader
than that.  It applies to the Work.  In the case of the Intel contribution,
it would clearly apply to ISORT, no matter where it is used.

>
>
> >
> > Where you get a potential divergence is in hypothetical cases where
> > people
> > are trying to game the system, taking a minimum amount of code
> > solely for
> > the purpose of getting access to the patent grant, but not taking
> > anything
> > closely approximating the intended "Work", not licensing the result
> > back
> > under the Apache license, etc.  While I appreciate Apache's interest
> > in
> > making sure that everyone can use its code, I don't think that the
> > team
> > should be worrying too much about unrealistic hypotheticals.
> >
> > So, does anyone have a likely use case where this would really be an
> > issue?
>
> Yes - any derivative work.  I still think that the scope of the patent
> grant is constrained to the specific Work to which the Contribution
> was made.
>
> (as I don't understand your argument that the patent grant applies to
> any apache licensed software)

In the spirit of getting specific, I'm proposing the following hypothetical
facts and provide my personal understanding of how the patent grant
interacts with the copyright grant in the Apache license.

 - Intel writes ISORT.
 - Intel has 3 patents: (A) one that read on ISORT itself, (B) one that
reads on ISORT when incorporated in a JRE, and (C) one that reads on ISORT
when incorporated in an App Server.
 - Intel contributes ISORT to Harmony.

Result, Intel grants a broad copyright right to do anything with the code
that Apache wants to.  Intel grants a license to use patent A because it
reads on ISORT itself and also to patent B because it reads on the
combination of the Contribution (ISORT) with the Work to which it was
contributed (Harmony).  The scope of the grant for those two licenses is to
deal in the "Work", which is a very flexible term and covers ISORT when you
are talking about patent A, as well as Harmony when you are talking about
patent B.

Anyone can download ISORT from Harmony and use it with the blessing of
Intel and coverage under patent A.  Whether they get coverage under patent
B depends on how much of Harmony they take.

New fact:

 - Someone takes ISORT from Harmony and contributes it to Geronimo.

This is just another example of dealing in the Work (ISORT).  Clearly,
we're covered under the copyright grant, and we are also clearly covered
under the patent grant to patent A.  We're also covered under the patent
grant to patent B, if it is infringed.  Not sure that it would be, but if
it is, the scope of the license covers it.  Do we get a license to patent
C?  Depends on who that someone was that copied the code to Geronimo.

 - That someone that copied ISORT to Geronimo was Intel

Then yes, Intel just recontributed the code to Geronimo and has provided a
patent grant to patent C.


Does this make sense to anyone else?
Jeff

>
> Thanks
>
> geir
>
>
> ---------------------------------------------------------------------
> DISCLAIMER: Discussions on this list are informational and educational
> only.  Statements made on this list are not privileged, do not
> constitute legal advice, and do not necessarily reflect the opinions
> and policies of the ASF.  See <http://www.apache.org/licenses/> for
> official ASF policies and documents.
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
> For additional commands, e-mail: legal-discuss-help@apache.org
>


Counsel, IBM Corporation  (914)766-1757  (tie)8-826  (fax) -8160
(notes) jthom@ibmus  (internet) jthom@us.ibm.com (home) jeff@beff.net
(web) http://www.beff.net/




---------------------------------------------------------------------
DISCLAIMER: Discussions on this list are informational and educational
only.  Statements made on this list are not privileged, do not
constitute legal advice, and do not necessarily reflect the opinions
and policies of the ASF.  See <http://www.apache.org/licenses/> for
official ASF policies and documents.
---------------------------------------------------------------------
To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
For additional commands, e-mail: legal-discuss-help@apache.org


Mime
View raw message