RE: draft-bonica-tunneltrace-02

[neil.2.harrison@bt.com]

Leen,

This is quite correct.  And it really does help not hinder.  The difficult
bit for most people I think is accepting MPLS fits here.  SDH and OTNs are
very obvious in that they do fit....ergo GMPLS fits.  FR and ATM also fit.
MPLS also fits but its hard for some to make this association (for whatever
reason....be it a truly held conviction or (more worryingly) commercially
driven motives).

All I have been saying is that this gives a formal way of describing layered
network relationships (just like your periodic table analogy)....and once
grasped it is very powerful in understanding/specifying required network
behaviour (and it also immediately explains the reality of the L1/2/3 tags
people use somewhat arbitrarily and the fact there there is no such thing as
'one' L3 network, viz the OSI 7-layer model).

It actually would help with progressing Bonnica and also understanding the
approach others have taken in Y.1710/1711.....but it requires a willingness
to be able to go into Rx mode rather than just Tx mode.

regards, Neil

> -----Original Message-----
> From: Mak, L (Leen) [mailto:lmak@lucent.com]
> Sent: 01 March 2002 15:58
> To: ccamp@ops.ietf.org
> Subject: RE: draft-bonica-tunneltrace-02 
> 
> 
> Let me make a Friday afternoon remark and than duck 
> for the flak:
> 
> IMHO, the layering concept (and the associated system
> of functional modelling) is to the study and design of
> (circuit based) transport networks what the Periodic 
> System of Elements is to chemistry.
> 
> Of course, you can do without, but with such an ordening 
> principle the work in the field becomes a bit less intuitive 
> and a bit more structured. If you like: less of an art and 
> more of an engineering discipline. 
> 
> I'm afraid this does not progress the discussion on the 
> bonica draft, but it might help getting the layering concept 
> where it belongs ...
> 
> 
> Leen Mak.
> 
>