Re: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt

[Dimitri.Papadimitriou@alcatel.be]

hi, to be cristal clear the problem is that what you 
indicate here below is the consequence of a decision
* consequence, which is a triviality by the way ie "if 
you don't have 2 solutions, you have a unique solution".

the decision was clearly related to the following issue
at the oif: how do i interoperate between rsvp-te and 
pnni at the signalling level and nsap vs ip end-point 
identification at the data plane level; the main issue 
is that *none* of the mechanisms currently defined or
considered by the oif gives a clear answer to these 
issues - but as said by bala, this proposal focuses on
rsvp signalling (ie no need for looking at interop 
issues with other signalling protocols *outside of
the box*) and uses ip/unnumbered for data plane only

thus your global statement is not valid with respect 
to its applicability scope

thanks,
- dimitri.

"Varma, Eve L (Eve)" wrote:
> 
> Hi John,
> 
> I don't want to go too many rounds on what was, after all, an
> OIF discussion.  However, there was a fairly sustained effort to
> establish p-UNI as a separate interface specification from
> several of our colleagues in the OIF.  It was however, decided, that it
> would NOT become a separate interface specification and the
> proposal to do otherwise was not accepted.  That is an accurate statement.
> 
> Best regards,
> Eve
> 
> 
> 
> -----Original Message-----
> From: John Drake [mailto:jdrake@calient.net]
> Sent: Thursday, December 12, 2002 3:42 PM
> To: 'Varma, Eve L (Eve)'; 'Stephen Trowbridge'
> Cc: 'Lazer, Monica A, ALASO'; Kireeti Kompella; Khuzema Pithewan;
> ccamp@ops.ietf.org
> Subject: RE: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt
>
> Let me try again.
> 
> I think it is a leap of faith to go from what the minutes say:
> 
> "It was agreed that further work related to Private-UNI/ILSI would take
> place under the auspices of UNI 2.0, and would not be a separate interface
> specification.  In particular, that services of interest be extracted from
> oif2002.540 and submitted as UNI 2.0 candidates (see candidate #19)."
> 
> to your interpretation of what the minutes say:
> 
> "Thus it is accurate to state that p-UNI/ILSI was not accepted (was
> rejected) as a separate interface specification."
> 
> -----Original Message-----
> From: Varma, Eve L (Eve) [mailto:evarma@lucent.com]
> Sent: Thursday, December 12, 2002 12:34 PM
> To: John Drake; 'Stephen Trowbridge'
> Cc: 'Lazer, Monica A, ALASO'; Kireeti Kompella; Khuzema Pithewan;
> ccamp@ops.ietf.org
> Subject: RE: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt
> 
> Hi John,
> 
> In fact, it was agreed within the OIF that the p-UNI/ILSI would not
> become a separate interface specification.  If you refer to the OIF minutes
> in
> OIF2002.269.4, you will see the statement:
> 
> "It was agreed that further work related to Private-UNI/ILSI would take
> place under the auspices of UNI 2.0, and would not be a separate interface
> specification.  In particular, that services of interest be extracted from
> oif2002.540 and submitted as UNI 2.0 candidates (see candidate #19)."
> 
> As no subsequent contributions related to possible additional services were
> submitted, candidate #19 was ultimately dropped.
> 
> Thus it is accurate to state that p-UNI/ILSI was not accepted (was rejected)
> as a separate interface specification.
> 
> Best regards,
> Eve
> 
> -----Original Message-----
> From: John Drake [mailto:jdrake@calient.net]
> Sent: Thursday, December 12, 2002 3:10 PM
> To: 'Stephen Trowbridge'
> Cc: 'Lazer, Monica A, ALASO'; Kireeti Kompella; Khuzema Pithewan;
> ccamp@ops.ietf.org
> Subject: RE: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt
> 
> I just noticed in this:  "Interface B (P-UNI/ILSI rejected by OIF,
> draft-swallow-gmpls-overlay-00.txt
> currently under consideration for WG status by ccamp)" the sly phrase
> characterization "rejected by OIF".
> 
> In point of fact, it was not rejected by the OIF.  This is the type of
> misinformation that I think Kireeti was talking about in an earlier e-mail.
> 
> -----Original Message-----
> From: Stephen Trowbridge [mailto:sjtrowbridge@lucent.com]
> Sent: Thursday, December 12, 2002 11:08 AM
> To: John Drake
> Cc: 'Lazer, Monica A, ALASO'; Kireeti Kompella; Khuzema Pithewan;
> ccamp@ops.ietf.org
> Subject: Re: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt
> 
> John,
> I think that the diversity of opinions indicates a couple of issues
> that are still open for debate.
> 
> There are two proposed interfaces:
> Interface A (using my terminology from the last email): OIF-UNI
> Interface B (P-UNI/ILSI rejected by OIF, draft-swallow-gmpls-overlay-00.txt
>   currently under consideration for WG status by ccamp)
> 
> I think there are carriers who want interface A who fear that standardizing
> interface B reduces their chance of getting A. Some work needs to go into
> alleviating that fear. One way is to get rid of interface B. Another might
> be to make it clear that vendors still intend to provide interface A
> instead of using interface B as a way to shortcut things and avoid giving
> (certain) carriers what they really want. For example, just because
> Microsoft sells an Office-XP "Home" edition doesn't prevent corporate
> users who really want it from buying Office-XP "Professional".
> 
> Another issue: are A and B different enough to be worth having two
> standardized
> interfaces? Are the shortcuts in B worth introducing interworking issues
> (will these ever be in the same network, e.g., at the enterprise/operator
> boundary?) and having to deal with those in the standards. Analyzing this
> is probably tricky with documents written in different styles produced by
> different organizations. Maybe we need to bring interface A (the OIF-UNI
> functionality) in as an Internet Draft so they can be looked at side by
> side.
> 
> Regards,
> Steve
> 
> John Drake wrote:
> >
> > Stephen,
> >
> > I couldn't help but notice that you snipped the part of my e-mail
> discussing
> > this topic, to wit:
> >
> > "JD:  If you're planning to assume the role of arbiter for the whole
> > industry, please review the e-mails all of who support accepting the
> > draft as a CCAMP WG document.  In particular I'd review the e-mails of
> > some of your peers:  Deborah Brungard (12/6), Gerry Ash (12/3) (both
> > of whom I think you know), as well as Mark Jones (12/2), and Jean-Louis
> > Le Roux (12/3)."
> >
> > The four individuals I explicitly mentioned work for carriers and the
> > e-mails I mentioned
> > all speak in favor of making the the draft a working group draft.  Also, a
> > representative
> > from another carrier is a co-author.  And lest we forget, two of the
> > individuals work for
> > the same carrier that Monica works for.
> >
> > Thanks,
> >
> > John
> >
> > -----Original Message-----
> > From: Stephen Trowbridge [mailto:sjtrowbridge@lucent.com]
> > Sent: Thursday, December 12, 2002 9:41 AM
> > To: John Drake
> > Cc: 'Lazer, Monica A, ALASO'; Kireeti Kompella; Khuzema Pithewan;
> > ccamp@ops.ietf.org
> > Subject: Re: WG Consensus Call: draft-swallow-gmpls-overlay-00.txt
> >
> > John,
> >
> > John Drake wrote:
> > (snipped)
> > > Snipped...
> > >
> > > >From a carrier perspective, supporting a third UNI alternative will
> > > bring additional concerns regarding interoperability and managing the
> > > network.
> > >
> > > JD:  No one is forcing you to supporting this interface, except perhaps
> > > potential customers.  To paraphrase George Swallow's 12/3 e-mail, if you
> > > don't think this interface is useful for you, please ignore it, but
> don't
> > > assume that it is not useful for others.
> >
> > Let's see, a carrier wants interface A.
> >
> > A vendor proposes to standardize another interface B which is similar, but
> > doesn't quite solve all the same problems as A, and tells the carrier "If
> > you don't like B you don't have to use it".
> >
> > It seems to me that the carrier concern is that if interface A and
> interface
> > B are both standardized, and if (the/some) carrier(s) want interface A and
> > (many/most) vendors choose to build only interface B, then the carriers
> > don't get what they want.
> >
> > This fear is what makes people reluctant to progress work on an interface
> > they don't feel is useful for them.
> >
> > The idea of standards is surely to promote the deployment of interoperable
> > implementations, but part of accomplishing this is to try to limit the
> > number of "standardized" solutions to the same problem.
> > I think the next debate would be what we really mean by "the same"
> problem.
> > Some arguments have appeared that this interface is directed at a
> > different problem, so I have to ask whether it is different enough
> > to justify standardizing a different solution.
> > Regards,
> > Steve

-- 
Papadimitriou Dimitri 
E-mail : dimitri.papadimitriou@alcatel.be 
Private: http://www.rc.bel.alcatel.be/~papadimd/index.html
E-mail : dpapadimitriou@psg.com
Public : http://psg.com/~dpapadimitriou/
Address: Fr. Wellesplein 1, B-2018 Antwerpen, Belgium
Phone  : Work: +32 3 2408491 - Home: +32 2 3434361