How to retrieve SP required attributes at the IDP before authentication

[Cantor, Scott]

On 3/21/12 3:22 PM, "David Chadwick" <d.w.chadwick at kent.ac.uk> wrote:
>
>this is rather disingenuous of you, given that you were not an author of
>the draft document, and were not proposed to be.

Now you're being disingenuous, because if that were true, you would not
have needed my input  or agreement at all.

> Coupled with the fact
>that prior to the draft that George Inman and myself wrote, an earlier
>one addressing the same topic and written by Sampo Kellomaki, was also
>turned down by the SAML group for standardisation, shows that there was
>some significant opposition to it.

That is not true. Sampo submitted an unfinished proposal, and then was no
longer actively pursuing it in the TC. His reasons are his own. Sorry, but
that is not "turned down" nor was there ever "opposition". In fact, my
recollection was that I noted the fact that multiple people had proposed
the idea and that they ought to talk about it so that a single proposal
would be worked on.

The SSTC has worked on a number of documents that I had no interest in, or
even disagreed with in some cases. As long as the documents were properly
written, correctly specified, and not a complete disaster, I have never
opposed them in any way. In fact, I've done more than my share of
commenting and correcting on such documents.

>So, whilst the ability to dynamically request attributes from an IDP
>along with the authentication request, has been an item of interest to
>several different researchers over the years, it has not been something
>that the OASIS group has wanted to support (to date).

That simply isn't true.

>Given the fact that you are not being asked to edit or shepherd the
>document, and the draft already exists, if it was re-submitted to the
>OASIS group, do you think there would still be strong opposition to it
>being standardised?

There never was any opposition at all, whatever you seem to think. But if
the submission is a mess with invalid XML, misuse of the schema, or other
deficiencies, or isn't in the right format, then no, it's not going to
just magically be accepted. Somebody has to do the work. In the TC. You
can't even submit anything if you're not part of the TC.

-- Scott