RE: Normatively referenced specifications

["Black, David" <[e-mail]>]

I definitely agree with Dave Crocker's "simple example" involving
IETF participation:

> Here's a simple example:
> 
>    Assume I own a patent on a component technology that has been
> published somewhere other than the IETF.  The organization publishing
> that specification did not require me to divulge my intellectual
> property claims.
> 
>    I now participate in an important IETF effort that considers
> including the component technology, by citing it normatively.  I of
> course, work vigorously to get the IETF to adopt the technology, but no
> one know that I stand to make serious money if the technology is included.
> 
> Given the intent behind the IETF's IPR rules, it makes no sense to allow
> me to participate without divulging my IPR interest in the topic.

For someone who participates in IETF in this fashion, I think RFC 3979 is
already clear on disclosure obligations - see sections 6.1.1 and 6.1.2,
which have been carried forward as sections 5.1.1 and 5.1.2 in
draft-bradner-rfc3979bis-06.txt .  I did not intend to contribute to calling
those obligations into question.

However, SM originally asked about "the licensing exercise", which is a
different concept:

> If, for example, an 
> implementation supports the XML format and XML is 
> covered by IPR, the implementation would have to 
> go through the licensing exercise.  From a 
> standardization perspective, would an IPR 
> disclosure be expected if the claim relates to a 
> normatively referenced specification?

For example, if the IPR in question were owned by an entity subject to
the W3C IPR policy, but that entity had no IETF participation, I would not
expect to see an IETF IPR disclosure.  In such a situation, the normative
reference to the XML spec would be the relevant clue to an implementer who
cares about applicable IPR - whatever happened wrt that IPR at W3C would
be the W3C's concern.

Admittedly, SM asked about this in the context of a quote referring to WG
discussions, so Dave Crocker's example was on point, but I think more
generally, for a spec developed outside the IETF, one cannot rely on
there being IETF IPR disclosures for all of the IPR that may have been
disclosed or otherwise known to whatever group developed that spec.

Thanks,
--David

> -----Original Message-----
> From: Dave Crocker [mailto:[e-mail]]
> Sent: Wednesday, December 18, 2013 12:40 PM
> To: Joel M. Halpern; Dale Mohlenhoff (dmohlenh); SM; Black, David
> Cc: [e-mail]
> Subject: Re: Normatively referenced specifications
> 
> On 12/18/2013 8:36 AM, Joel M. Halpern wrote:
> > Dale, your interpretation is exactly the opposite of what I thought
> > Dave was saying. Referencing a spec normally does not and ought not
> > mean that the referencers IPR rules apply to the referenced
> > specification.  They can't, since in the abstract the referencer has
> > no control over the IPR grants of the referenced document.
> >
> > And IETF IPR grants apply when implementing IETF RFCs, even when
> > those RFCs are implemented in conjunction with some other
> > specification.  The IPR grants don't apply (as I understand it all
> > bets are off) if you modify the spec.
> >
> > Equally, the IETF can not insist that our IPR rules apply to a spec
> > we reference.
> 
> 
> I think this confuses the nature of what our disclosure rules mean. For
> example, they don't mean that we are affecting some other standards
> group.  Rather they affect decision-making within the IETF.
> 
> So, absent a clear consensus from legal experts who are familiar with
> the IETF's IPR rules, I disagree with Joel's assessment.
> 
> 
> As I just posted on rtcweb:
> 
> > Normative language in a specification defines the syntax and
> > semantics of the thing being specified.
> >
> > It does not make much sense to handle IPR differently for normative
> > text that includes details by reference (citation) rather than by
> > inline explication.  In terms of the syntax and semantics, the
> > specification is an integrated whole.
> >
> > If there is an legal distinction between inline normative reference,
> > versus citation-based inclusion, which is relevant to the
> > interpretation of IETF IPR rules, then some lawyers should provide
> > the community with expert opinions on the matter.
> >
> > Absent a clear and compelling presentation from legal experts, common
> > sense needs to prevail in the IETF's interpretation of its rules.
> >
> > Our rule is quite simple:  participation obligates disclosure.
> 
> 
> Here's a simple example:
> 
>    Assume I own a patent on a component technology that has been
> published somewhere other than the IETF.  The organization publishing
> that specification did not require me to divulge my intellectual
> property claims.
> 
>    I now participate in an important IETF effort that considers
> including the component technology, by citing it normatively.  I of
> course, work vigorously to get the IETF to adopt the technology, but no
> one know that I stand to make serious money if the technology is included.
> 
> Given the intent behind the IETF's IPR rules, it makes no sense to allow
> me to participate without divulging my IPR interest in the topic.
> 
> d/
> --
> Dave Crocker
> Brandenburg InternetWorking
> bbiw.net
> 
> --
> Dave Crocker
> Brandenburg InternetWorking
> bbiw.net