Re: IESG Evaluation of draft-ietf-pppext-vendor-protocol-01.txt

Jari Arkko writes:
> > 	- Making 2153 an informative reference. (doable, but would
> >           require some text shuffling)
> 
> Hmm.... perhaps you can do it but I wonder if its too
> much of a stretch. The draft says that it uses the
> 2153 OUI values. Yes, they are mostly IEEE numbers
> but not in the case of CF0000 values.

I don't think that's so much a problem -- this draft can reference
IANA for those values directly.  The CF0000 values are fixed by IANA,
regardless of what 2153 might say.

> > 	- Moving this draft to Informational track.
> 
> I'm not sure why 2153 was put out as Informational.

I *think* the reasoning looked something like this:

	These are, by definition, vendor-proprietary extensions.
	There's no hope that vendors will interoperate on proprietary
	extensions (if they did, such things would be better issued as
	public documents), so any document that depends on such
	interoperability won't advance on standards track.

The small flaw behind that logic, at least in my mind, is that
properly agreeing-to-disagree (per the protocol, where each side is
from a different vendor) *is* an instance of "interoperability," so
long as the link doesn't fail or behave badly because of it.

There's a fuzzy line here -- the mechanism for extension is itself
reasonably standards-track material, but any use of the mechanism
would (necessarily) be proprietary.  So what do you do?

All that said, I'm not too worried about the problem.  If the simplest
solution is to make this Informational as well (and I think it is),
then that's certainly fine by me.  It's publication in a fixed form
that allows reference ("here's what you do if you're doing one of
these") that matters most to me.

-- 
James Carlson, IP Systems Group                <[email protected]>
Sun Microsystems / 1 Network Drive         71.234W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.497N   Fax +1 781 442 1677