www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Thomas <ma...@apache.org>
Subject Re: Proposal: Disclosure of patents by Apache projects
Date Wed, 03 Jun 2015 14:42:54 GMT
On 03/06/2015 14:38, William A Rowe Jr wrote:
> On Wed, Jun 3, 2015 at 7:59 AM, Shane Curcuru <asf@shanecurcuru.org
> <mailto:asf@shanecurcuru.org>> wrote:
> 
>     On 6/3/15 8:44 AM, Geir Magnusson Jr. wrote:
>     > But what's the point of any of this?  There's no reason to require Sneaky Author
to enumerate anything because via contribution via the AL, they have granted a license to
anything they have necessarily infringed by the contribution (or combined contribution and
existing work)
>     >
>     > And I as the recipient don't really care - I know that via the patent grant
in the AL, Sneaky Author can't come after me for any patents they hold (enumerated or not)
that read on the work they contributed to.
>     >
>     > Is there a real problem we're trying to solve?
> 
>     I'm not aware of any Apache project asking this question, no.  It seems
>     Bill might have a case related to this thread, which we can address when
>     it comes up.
> 
> 
> Indeed, there is no sneaky author.  The patents are public filings.  No 
> need for tinfoil hats or detection gadgets.  Those patent use rights are 
> granted, by the very design of the AL, to every downstream consumer.
> 
> Apparently, none of you spend too much time with your corp-consumer
> legal teams.  It's brutal.
> 
> Pretty much everything out there, IP-wise, bollixes the works.  Stray
> non-conforming license?  That's out.  Stray patent claim?  Straight out.
> 
> So yes, there is a project which is trying to offer code-with-patent-license
> to the ASF, and has succeeded at their CCLA plus CLAs and would like
> to wokr out the best way to inform / reassure users that these particular
> patent claims are known to apply and are granted for use to whatever
> downstream consumer wants to use the code.

<full-disclosure>
The corp in question is my employer.
These are my opinions not my employers.
</full-disclosure>

I appreciate that the corp in question is trying to do the right thing.

I would argue - based on our experience of adding the "I grant this
patch under the ALv2" button to Jira - that doing anything over any
beyond that which is required by our existing CLA and CCLA process will
cause us more problems in the long run than it solves.

By granting this code to the ASF under the ALv2 any and all necessary
patent licenses are granted. That is the only thing that needs to be
said publicly and - in my view - the only thing that should be said
publicly. There is no problem here that needs to be solved.

If the corp wants to provide the PMC (privately) with a list of patents
it thinks it has licensed then fine but I don't see how making that list
public helps anyone.

The potential issues I see with making the list public include:
- projects that don't publish a list of licensed patents start being
  asked by users to produce one
- projects that do publish a list start being asked about XYZ patent
  that a random user things might apply to the project
- committers start being asked to explicitly state if each commit is
  covered by a patent or not

If we could guarantee that - for each project - we could produce a
complete list of licensed patents then I'd have far fewer concerns. But
I do not believe we can do that. Further, I believe that publishing
incomplete lists will create the perception of problems where none
exists given the clear and unambiguous language that is already present
in the ALv2 with respect to patents.

Mark


---------------------------------------------------------------------
To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
For additional commands, e-mail: legal-discuss-help@apache.org


Mime
View raw message