www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael Chen Lee" <mchen...@yahoo-inc.com>
Subject RE: CCLA: "by combination" language
Date Wed, 09 May 2007 20:00:54 GMT
Hi All: 

I'm new to the legal discussion group, and wanted to pick up on an old
but important topic that still is in controversy - what Jeff Thompson
describes in the below thread as the "horizontal" scope of a "Work" in
relation to the Apache CLAs.  If one submits a contribution to an Apache
project or sub-project, is that contribution considered to be submitted
to *only* the specific project/subproject at hand or to *all* ASF
projects/subprojects?  If the latter, are the patent claims licensed
under the CLA those claims necessarily infringed by the combination of
one's contribution with (a) *only* an intended project/subproject at the
time the contribution was made, or (b) *all* ASF projects/subprojects?  
  
It seems to me that interpretation (a) is the only reasonable option, as
it gives contributors the clearest and most reasonable scope as to which
patent assets it may be licensing, when it intends for its contribution
to be used in a specific subproject by submitting its contribution to
that subproject.  

Interpretation (b) could mean that all of one's patent claims triggered
by the combination of one's contribution with any and all Apache
projects would be licensed, regardless of the specific subproject to
which that contribution was explicitly submitted.  This result, to me,
is unreasonable and could not have been what the original Apache CLA
drafters had intended. 
  
However, I know that others on this thread have taken the opposite view
- namely, that if the definition of "Work" is broad, the resulting
patent license must also be broad.  For example, see Jennifer O'Neill's
5/26/06 post ("...the intent of the working group was that once a
Contribution was "intentionally submitted" to ASF for inclusion in the
open source community, any of the ASF projects could be the "Work," so
that the patent grant broadly covered the combination of the
Contribution with any conceivably relevant ASF projects.").   
  
There seems to be a fundamental disagreement on this aspect of the CLAs'
patent license provision.  Does Apache have a formal position on this?
Given the differing opinions on this so far, I think Apache contributors
and the community at large would benefit by having clarity on this key
point.  Thanks in advance and look forward to your collective thoughts.

Mike

_________________________________
michael chen lee  |  yahoo! inc.
sr. legal director, intellectual property
tel:  408.349.2829  |  fax:  408.349.3400
mchenlee@yahoo-inc.com



Message view < Date > * < Thread > 
Top < Date > * < Thread > 
>From Jeffrey Thompson <jt...@us.ibm.com> 
Subject RE: CCLA: "by combination" language 
Date Thu, 25 May 2006 19:04:05 GMT 
Jim, Roy, Cliff, et al.,
   OK, I'll throw my 2 cents in as well, since I agree with Jim's
initial 
notes and disagree with Roy's position.  Below is the relevant language
as 
provided by Jim and my thoughts on what it means and why its there . . .
.

> "Subject to the terms and conditions of this Agreement, You hereby
grant
> to the Foundation and to recipients of software distributed by the
> Foundation 

The grant is to Apache and directly from the patent holder to any 
recipients of the Apache software . . . 

> a perpetual, worldwide, non-exclusive, no-charge,
> royalty-free, irrevocable (except as stated in this section) patent
> license 

nothing surprising here

> to make, have made, use, offer to sell, sell, import, and
> otherwise transfer the Work,

The patent licenses authorize the licensee to do things with the Work. 
There are 2 questions as to what "Work" means.  The current question, 
which is "temporal" -- does the Work include future versions of the
Work. 
Also, there is also a "horizontal" question -- whether the Work only
means 
that specific piece of code in the specific Apache project to which the 
Contribution was initially made.  Since the copyright license permits
any 
Contribution to be used in any of the Apache projects, one could take a 
Contribution in one project and use it in another.  So, if you make a 
Contribution to HTTP Server 2.2.2, the first question determines whether

your patent license covers HTTP Server 2.3 and the second determines 
whether your patent license covers Tomcat (assuming your Contribution
gets 
used in the Tomcat project).  I think that its clear that Apache intends

that the temporal issue is covered by the term "Work".  I also think
that 
the normal use of the term "Work" covers the horizontal issue as well. 
Looking closely at the definitions, once your Contribution is included
in 
Tomcat, you are a "Contributor" per the definitions as it relates to 
Tomcat, and your patent grants would cover Tomcat as well.

> where such license applies only to those
> patent claims licensable by You that are necessarily infringed by Your
> Contribution(s) alone or by combination of Your Contribution(s) with
the
> Work to which such Contribution(s) were submitted."

As Roy points out, this last part selects which patents to which the 
license applies.  Patents that are infringed by the Contribution itself 
are clearly licensed.  In addition, if there is a combination, it covers

patents infringed by the Contribution in combination "with the Work to 
which such Contribution(s) were submitted".  The open question is
whether 
that extra qualifier "to which . . ." intended to address the temporal 
question or the horizontal question (as I named them above) or possibly 
both.  I  had thought that this clause was intended to address the 
horizontal question, that is, the patent selection is determined by the 
HTTP Server (as it evolves), but if someone takes your Contribution and 
copies it over to the Tomcat project, any additional combinations
possible 
because of the extra scope of the Tomcat project are irrelevant.  That 
makes sense to me and with that interpretation, the language isn't 
meaningless.

If, however, the Apache consensus is that we think that the add'l
language 
should also limit the Work temporally, we should make that clear in a
FAQ. 
 Come to think of it, since this has come up as an issue, we should 
probably make whatever we decide clear in a FAQ anyway. 

Jeff 

Staff 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