Re: [security-services] Groups - Draft Webinar on SAML 2.1 Plans uploaded

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

On 7/23/12 5:02 PM, "Hal Lockhart" <hal.lockhart@oracle.com> wrote:

>>Does anyone have any raw data or general characterizations I can use? I
>>will be glad to work it up into slides, but I simply don't know anything
>>precise I can say. I know there are "many" deployments and some are
>>"very large" but other than certain specific ones, like Internet2 I
>>don't know specifics.

For higher ed, there's this table:
https://refeds.terena.org/index.php/FederationsTable

Most of those (all?) are SAML based federations at this stage, at least in
part.

I would say that in terms of "characterizing" deployments, what we have is
a large, unknowable number of the usual point to point stuff that people
think SAML was designed for, and then we have, in higher ed, and in a few
other vertical deployments, a lot of very expansive federations that are
multi-lateral and are meant to scale very heavily. Most of the post-2.0
work that the slides cover was written by me or my colleagues and was
geared to those scenarios.

>Personally I have to refer to the summaries just to remember what they
>are about. I found the entire list pretty mind numbing. However if others
>agree with Chad, I will try to compress them into 4-5 slides.

I think it's better maybe to cover just a handful that are more "key" to
motivating 2.1 work, at least motivating me. That set is probably the
discovery and request inititation stuff in one bucket, which addresses
discovery (duh) and the security issues around IdP-initiated, and the
other is all the metadata extension and profile work.

Some misc. comments:

On slide 5 under Metadata, can I suggest s/agreements/connectivity?

I think most of the Uncommitted Work bullets are going to be lacking a
work champion, so I'm wondering if it's best just to link to the wiki
rather than enumerate them. Maybe it's worth just saying "soliciting
proposals for additional work items from parties willing to contibute to
them".

-- Scott