Re: [TLS] Straw poll on TLS SRP status

[Paul Hoffman <paul.hoffman at vpnc.org>]

At 2:05 PM -0700 5/24/07, Nelson B Bolyard wrote:
>  >    [x] Something else, please state:
>
> I think the WG must be consistent about whether or not IPR (other than
> royalty-free IPR) forces an RFC into Informational/Experimental status.
> That decision should not be revisited for each and every RFC that may
> encounter it.
>
> RFC 4492, the ECC-in-TLS RFC, was forced into Informational because the
> IPR status was uncertain to some WG members, even though there are two
> major browsers that offer it (one open source), servers from several
> vendors that offer it, and it is implemented in at least TWO different
> open source TLS implementations.  I think that's sufficient precedent.
>
> Note that I don't think that "Informational/Experimental is better."
> but rather that it is what the WG has done in the past, and the WG
> should remain consistent, or else the chairs should explain why they
> insisted on it before and are now not doing so.

Fully agree. And I think it is appropriate to delay the decision on 
tls-srp until we have made that decision. Such a delay will not 
affect developers significantly.

FWIW, I'm fine with us saying "they can get on standards track just 
as well as if there were no IPR issues", but I'm equally fine with 
"they can only be published as Informational".


At 4:02 PM +0300 5/25/07, <Pasi.Eronen at nokia.com> wrote:
> The decision cannot depend only on the existence of IPR claims. 
> Not all IPR claims and license terms are equal.  And IPR claims
> anyway usually cover only IPR held by those contributing to the
> work; it's even likely that other IPR might be found later,
> after the document has been published.

If more than a small handful of TLS WG members think that a 
properly-published IPR claim is likely to be valid enough to be held 
against them, that is a high enough bar to put the proposal in the 
"IPR-affected" bucket.

> But most importantly, not all documents are equal. There are some
> things we probably want on standards track even though there is
> other-than-royalty-free IPR (IPsec is one obvious example).

This is irrelevant. We are talking about TLS extensions, not TLS itself.

> Other
> documents could be better off Informational/Experimental even in the
> absence of any IPR.

This is also irrelevant.

--Paul Hoffman, Director
--VPN Consortium

_______________________________________________
TLS mailing list
TLS at lists.ietf.org
https://www1.ietf.org/mailman/listinfo/tls