RE: patent licenses on OASIS standards

[Jeffrey Thompson]

"Lawrence Rosen" <[EMAIL PROTECTED]> wrote on 02/25/2005 11:44:31 AM:
> Hi Jeffrey,
>  
> My proposed Open Standards Principles are completely consistent with
> the W3C patent policy, with the addition of sublicensing. I invite
> you to compare them. I also invite you to compare it with the IBM
> non-assert statement of a few weeks ago. Except that IBM’s non-
> assert is only for open source implementations rather than to
> “everyone,” it too is consistent. I think you’ll find that’s true
> also for most of the licensing commitments for standards that we
> already think of as adequate for our needs. So my definition isn’t
> very far off, I believe.
>  

I don't see them as consistent, but we'll defer that discussion to another day.

> I want to separate out the discussion of which specific definition
> of open standard is adopted from the notion that we should have some
> such definition and we should enforce it. I responded yesterday to
> Sam Ruby’s email with my proposed definition, but I wasn’t trying to
> force it down everyone’s throat. There’s room for modification or
> even the adoption of something different. The problem so far is that
> there hasn’t been a proper venue for that discussion within
> standards organizations because of the enormous shadow of RAND.
>  

I agree that Apache certainly needs to understand the effect of its choices to implement standards.  Creating a filter to identify those standards which Apache projects should steer clear of would go a long way toward avoiding nasty surprises later.   That seems like a good project to work on.  Whether that should be translated into a broader OSS-wide definition of "open standards" is another question.  I don't know how easy that would be and once its done I don't know how it would be used?  Who is going to go tell SAMBA to stop implementing CIFS, for example?  

In any event, Apache guidelines would address, in my mind, at least three basic questions:

1.  Can Apache get the standard?  If we can't get it, we can't implement it.

2.  Can Apache publish its implementation under Apache's license?  This is the most critical.  Any standards agreement that prevents open source implementations shouldn't be embraced by Apache.

3.  Is Apache opening itself or its customers to royalties for necessary patents?  This is the hardest to answer.  Your definition of open standards spent a lot of time talking about the details of the patent licenses, but in the end, the question is whether the open source project and its customers qualify for the free license.  Furthermore, the holders of all necessary patents might not be known.  

In the end, it boils down to is there a known patent holder that has refused to offer an "acceptable" RF license for necessary patent claims.  Even for standards that are nominally published under RAND commitments, if all of the known patent holders have made RF commitments that cover Apache and its customers then there is no practical roadblock for Apache's open source implementation.

> Regards,
>  
> /Larry
>  
> Lawrence Rosen
> Rosenlaw & Einschlag, technology law offices (www.rosenlaw.com)
> 3001 King Ranch Road, Ukiah, CA 95482
> 707-485-1242  ●  fax: 707-485-1243
> Author of “Open Source Licensing: Software Freedom
>                and Intellectual Property Law” (Prentice Hall 2004)
>  

Jeff


Staff Counsel, IBM Corporation  (914)766-1757  (tie)8-826  (fax) -8160
(notes) [EMAIL PROTECTED]  (internet) [EMAIL PROTECTED] (home) [EMAIL PROTECTED]
(web) http://www.beff.net/