The MPLS WG Archive

["Hamid Ould-Brahim" <hbrahim@nortel.com>]

Hi Stewart, 

> 
> Hamid Ould-Brahim wrote:
> > Italo,
> >
> >   
> >> I feel a little bit unconfortable with the second part of the 
> >> proposed liaison:
> >>
> >> "The PWE3 WG will perform a further review of G.8110.1 Amendment 1 
> >> and will respond in time for the ITU SG15 meeting in June."
> >>
> >> It sounds like the draft has not been read at all. I do not think 
> >> this is a polite message we wish to give to ITU.
> >>
> >> I would propose to smoote a little bit the message with something 
> >> like:
> >>
> >> "The PWE3 WG has not found other major issues than those 
> identified 
> >> by the MFA Forum, however it will perform a further 
> detailed review 
> >> of G.8110.1 Amendment 1 and will respond in time for the ITU SG15 
> >> meeting in June."
> >>
> >>     
> >
> > Since PWE3 intends to perform a detailed review of the proposal I 
> > don't think at this point in time PWE3 is in position to 
> conclude that 
> > it "has not found other major issues than those identified 
> by the MFA 
> > forum".
> >
> > Hamid.
> >   
> Thanks Hamid, absence of proof is not proof of absence.
> 
> Please note that the ITU (T)MPLS work has been taking huge 
> resources from the PWE3 & MPLS ADs and chairs. Since last 
> summer one week of 1AD + 2WG  chairs, and one week of 2WG 
> chairs  to ensure that the first version of T-MPLS  complied 
> with the IETF PWE3/MPLS architecture - and when we did find 
> an error it took significant time to negotiate the changes 
> needed for compliance in multiple ITU groups across multiple 
> documents.
> 

I think one of the issue I see from the thread discussions
is it looks that MPLS and PWE3 WG (IETF) are not comfortable 
with the use of MPLS acronym (and technology) in 
ITU <T>-MPLS work. 

Though the (G)MPLS change process did actually define 
the process of extending MPLS by other SDO, it did not 
address the case where other SDOs standardize technologies 
that use some of IETF MPLS-based documents and then label 
or identify these technologies as being "<X>-MPLS-based".

I think it is up to the MPLS WG (and IETF in general) to 
evaluate whether such use of MPLS is *okay* (using the
change process) and is as well *okay* to call the proposed 
technology <X>-MPLS. 

It looks from the T-MPLS discussion here that it looks
not enough that a proposal uses an IETF defined MPLS label, 
removes IP, turns off some MPLS features and adds new non 
IETF MPLS/PWE3 capabilities for the end result to be qualified 
as <X>-MPLS. This creates additional confusions (on what version
of MPLS each SDO means) and leads to all these 
lengthy time consuming liaison processes just to get common
understanding of the intent and compliancy to IETF MPLS
architecture/RFCs. 

On the other hand, with respect to using MPLS in
transport networks for example, there is actually no work 
in MPLS working group that clarifies how MPLS works 
in Non-IP environment. 

I think MPLS WG should describe in an internet-draft
the "Application of MPLS in Non-IP Network 
Environment" (I assume some of it is already captured in Stewart 
pwe3-transport draft). And Stewart's draft should refer to 
that work too.

That draft can be liaised to other SDOs that intend to use MPLS 
in non IP environment (along with the change process document).  
 
Hamid.

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls