RE: Loops (RE: CPIM changes)

> -----Original Message-----
> From: Dave Crocker [mailto:[email protected]]
> >> -----Original Message-----
> >> From: Dave Crocker [mailto:[email protected]]
> >>
> >> The group of nine was commissioned to specify a common 
> >> gateway semantic. That is all. Not a protocol. Not a content
> >> syntax. But rather the minimum acceptable gateway
> >> functionality that could pass muster among the group of nine.
> 
> Adam> Huh. Derek's presentation to IPMM at the time seemed to
> Adam> indicate otherwise.
> 
> Whereas I think that the Introductory language in the 
> original CPIM document
> is straightforward and clear:
> ----- [ ]
> The parameters for each operation are defined using an abstract
> syntax. Although the syntax specifies the range of possible 
> data values,
> each IM service must specify how well-formed instances of the abstract
> representation are encoded as a concrete series of bits.
> -----

In context [1], that passage is followed immediately by an example:

  "For example, one strategy might transmit presence information as
   key/value pairs, another might use a compact binary representation,
   and a third might use nested containers. The choice of strategy is a
   local matter, providing that there is a clear relation between the
   abstract syntax (as specified in this memo) and how it is faithfully
   encoded by an particular IM service."

The "clear relation between the abstract syntax... and how it is
faithfully encoded" sure sounds like a lossless syntactic
transformation to me.

> Over time, folks have forgotten the intended goal and have,
> instead, turned this into yet another effort to do an
> end-to-end common im&p protocol.

Okay, so if these fine details of purpose have been lost
lost in the institutional memory of the IMPP working group,
then consensus polling from that period [2], when the relevant
events were still fresh in everyone's minds, should
roughly agree with your assertions, right?

  Q: do we think that we need to be able to pass around a pile
     of bits that can be signed:
  A: Rough consensus in the room (all but 1 raised hand yes)

  Q: Should CPIM specify structure of this message?
  A: All think it should specify the format.

Huh. Odd.

/a

[1] http://www.ietf.org/proceedings/00dec/I-D/draft-ietf-impp-cpim-00.txt
    (I'm citing the same document as the original reference, although
     the change between the original and the current document appears
     to be solely editorial).

[2] http://www.ietf.org/proceedings/00dec/00dec-16.htm



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