Re: Section 6.5: Additional Licenses for IETF Contributions

[Simon Josefsson <simon at josefsson.org>]

On 7 maj 2007, at 20.07, Harald Alvestrand wrote:

> Simon Josefsson wrote:
> > "Joel M. Halpern" <joel at stevecrocker.com> writes:
> >
> >
> > > b) With regard to code that is licensed under different rules with
> > > some restrictions not in the IETF rules, that code can not be  
> > > included
> > > as sample code in the IETF documents.  I was told that the rough
> > > consensus was against allowing more restrictive licenses to be
> > > included in the documents.  So, for example, any code that requires
> > > that changes be returned to the origin can not be included as  
> > > code in
> > > an IETF document according to these rules.
> >
> > That would make it impossible to include example code licensed under,
> > e.g., the BSD, MIT or similar licenses, in IETF document.  Doing that
> > has been done with good results in several RFCs.  It can be argued  
> > that
> > including example code, under various licenses (including less
> > permissive than this, which there are many examples of in past RFCs),
> > furthers the goal of the IETF, and therefor should be permitted.  The
> > argument against appear to be variations on the following theme:
> The argument against is that it's onerous and nearly impossible for  
> users to track down the provenance and licensing of each individual  
> piece of code in an RFC to verify its usability, unless the IETF  
> can help them.

A natural requirement to publish something as an IETF document would  
be that any copyright notices and licenses are clearly marked.

Btw, having clear markings is what the current WG documents imply  
too: documents need to clarify which parts are licensed under the  
future "IETF code license" and which parts are under the "IETF text  
license".

> The conclusion of the WG, as I understood it, was that we wanted  
> implementors to be able to exercise the set of rights that the IETF  
> grants to users of IETF code to all users of all IETF code (going  
> forward from the day of implementation).

I understood it as a conclusion to not permit copyleft licenses, but  
nothing further.

> > > Would it be nice to be able to use such code?  Sure.  But having
> > > multiple different sets of rules, and license terms in documents,  
> > > and
> > > efforts to decide ~are we willing to include this code with this
> > > license?~ was not what the group decided.
> >
> > This situation would be the same as for patents.
> No, because the publication of a text (such as an RFC, or a  
> citation from an RFC) can never violate a patent.
> The publication of a text can violate a copyright.

Handling a problem like that is why we write policies that  
contributors must follow.

/Simon


_______________________________________________
Ipr-wg mailing list
Ipr-wg at ietf.org
https://www1.ietf.org/mailman/listinfo/ipr-wg