license with patent grants appropriate for specifications?

[todd glassey]

The problem is ultimately that one size doesn't fit all needs, and that
means either reconfigurable licensing or multiple licenses.

Todd

----- Original Message ----- 
From: "Bob Scheifler" <Bob.Scheifler at Sun.COM>
To: <lrosen at rosenlaw.com>
Cc: <license-discuss at opensource.org>; "'Ipr-Wg at Ietf.Org'" <ipr-wg at ietf.org>
Sent: Wednesday, January 12, 2005 7:13 AM
Subject: Re: license with patent grants appropriate for specifications?


> > Most robust open source licenses contain both copyright grants and
patent
> > grants. These licenses should be sufficient.
>
> This goes to the heart of my original question. To be concrete,
> if use the Apache License, Version 2.0 for a specification, and
> I look at its grant of patent license, it covers patent claims
> that are "necessarily infringed by the [specification]", it does
> not explicitly cover patent claims that are necessarily infringed
> by implementations of the specification. Am I correct in assuming
> that a mere specification cannot directly infringe a patent claim?
> (It may well be subject to claims of contributory infringement.)
> In which case, the grant of patent license would seem to be
> insufficient for enabling implementations. Or is there some other
> existing open source license with more suitable grants? Looking at
> AFL 2.1, perhaps "patent claims ... embodied in the [specification]"
> is broad enough, but since the grant is only to "make, use, ... the
> [specification] and Derivative Works", is that reasonably interpreted
> as extending to implementations of the specification? And the AFL 2.1
> termination provision appears to be limited to claims that the
> "[specification] infringes a patent", so the same question about
> whether a specification can directly infringe a patent arises.
>
> - Bob
>
> _______________________________________________
> Ipr-wg mailing list
> Ipr-wg at ietf.org
> https://www1.ietf.org/mailman/listinfo/ipr-wg