RE: References to MLPP that works Re: I-D ACTION:draft-ietf-ieprep-domain-req-00.txt

Steve,

Just a couple of comments for clarification to hopefully alleviate some
of your concerns...

> If a call/session is in progress between a non authorised ETS user and
a
> emergency service then this should never be subject to pre-emption by
a
> authorised ETS user, it could have fatal consequences. This by
extension
> could be applied to a call/session in progress between non authorised
> ETS users, as this contact could be the only point of hope for someone
> in trouble.

In the soon-to-be-released RFC-3689, General Requirements for ETS, there
are a couple of pertinent excerpts about "policy".  Paraphrased, it says
that a set of labels may exist and that they may have a stated
relationship to each other.  In the case of MLPP, this would include
preemption. HOWEVER, that relationship is only for *that* set.  This
means that the characteristic of preemption does not automatically
correlate to a different set of defined labels.  Instead, that
relationship is defined by local policy.

Put another way, the fact that MLPP has precedence does not mean that
the same characteristic automatically applies to other labels (ie, other
ETS type labels) or non labeled traffic.  Instead, you the operator (if
you support ETS labels) decide the relationship between different sets
of ETS type labels, as well as the relationship between ETS and non-ETS
labeled traffic.  If this is still not clear, contact me privately and I
will expand.

Note: In the case of BT, Oftel may make that decision for you based on
laws/regulations, but that's outside the scope of the technology.  In
the case of the US PSTN, automated preemption is not allowed.  

-ken

ps, the predominance of MLPP systems in use today are in private
(non-PSTN) systems.