RE: [security-services] specification text question and editorial nits

["Scott Cantor" <cantor.2@osu.edu>]

> I am not sure what the role of this text is. Is it an 
> implementation hint? I am specially puzzled by the last 
> sentence --- I don't see that it is implied by the remainder 
> of the specification text. 

It's a clarification required by Dave McAlpin who asked why an assertion can
be used as an encrypted identifier. It's not a hint, it's a direct statement
about the ability to do this and the semantics when you do it. The last
sentence is an additional processing rule that is an extension of the use
case itself...to limit the use of an identifier in time or space. If you
were allowed to ignore the validity of the identifying assertion in your
evaluation of the containing assertion, there would be no value in even
bothering to permit this.

> I propose the following editorial changes: 
> (1) R/sending of the message/transmission of the message/
>  (2) R/sender acts as an intermediary/presenter acts as an 
> intermediary/

Agree, especially the second. I think that's a screw up on my part back when
the name of this actor changed.

-- Scott