RE: AW: Any resolution to the QSPEC mandatory parameter li st?

John,

> I think we might want to think about creating a generic QoS
> model that allows endpoints to signal only what parameters
> are needed. 

Not sure I get what you mean.  A QNI can already signal 'only what
parameters are needed'.  E.g., there is no need to signal all mandatory
parameters, some or none can be signaled, for example (that's been a
point of confusion).

Jerry