RE: Communication means in draft-ietf-impp-cpim-pidf-05

> > RFC 2778 isn't trying to dictate a particular model of what a tuple
> > might represent. Some systems might associate tuples with devices.
> > Others, with users. Others, with identities. The point of 2778 is to
> > provide a general framework which can be used to describe a variety of
> > presence systems, not to dictate the structure of a particular system.
>
> So, why use tuples instead of, say, MIME multipart with a different PIDF
> document in each of the multiple parts? Or different status
> elements in the
> same tuple?

I think there are a couple of different answers here depending on the
question you're asking.

One such question is "why did 2778 use tuples?" The answer is that it
worked, and the alternatives that you offer were unacceptable as a
description of the multiple systems to be described. For example, if your
favorite presence system doesn't use MIME multipart to carry presence info,
then a supposedly general descriptive framework relying on MIME multipart
has a bug.  Tuples work as an abstract description of MIME multipart and
things like it, but not vice-versa.

The other such question is "why does IMPP use 2778 to define CPIM?" That is
the more difficult question to answer. I think the 2778 authors have been
among the most reluctant participants in the group's use of 2778 as a
prescriptive mechanism, since they know perfectly well that 2778 starts out
by saying you shouldn't do that. But it's hard to know what to do instead.
The IMPP charter is effectively to achieve some level of interoperability
among groups that don't otherwise agree on much beyond 2778/2779.

--Mark




  [reminder: [email protected] for non-technical discussions, please]