References

[Evan Prodromou <evan@outlook.net>]

Well, I'm looking over the ballot comments on references, and I have
to say that I'm a little disappointed.

I really think that there is a strong requirement for some kind of
abbreviated form of "assertion handle" -- a way to get a full
assertion from a small bit of data. XML is wordy, and we need small
data for browser bindings.

This is a non-obvious requirement for SAML that flows out of the use
cases in our doc. I think it should be in there. Fortunately, since we
didn't get consensus, R-Reference stays in, unclear as it may be.

I was especially a little bummed at ballot comments that said, in
response to the ballot issue to further define references,
"'Reference' is undefined." Yeah, that's why we're trying to define
it!

~ESP