Interfederation

[Tom Scavo <>]

On Tue, Apr 30, 2013 at 4:53 PM, Cantor, Scott 
<>
 wrote:
> On 4/30/13 4:47 PM, "Tom Scavo" 
> <>
>  wrote:
>
>>On Tue, Apr 30, 2013 at 3:20 PM, Basney, Jim 
>><>
>> wrote:
>>>
>>> Need to separately discuss SP and IdP.
>>> Could use R&S category on IdP side in short term.
>>> Tackle IdP discoverability in long-term.
>>
>>Scott, would it be possible for you to expand on this a little bit?
>
> I was saying that the issues around opt-in/out are vastly different for
> IdPs and SPs

Totally agree.

> (e.g. I see no support issues around exposing non-federated
> SPs to new IdPs that won't care and won't be offered to users anyway).

Perhaps, but I don't see much point exporting 1100 SPs, the vast
majority of which are non-federating. The best estimate we have of
federating SPs is R&S, that is, 12 SPs. If there are more, they should
be applying for R&S (which has a fairly low bar).

Turning this around, consuming SP metadata is problematic because
imported SPs will not be bound by our Participation Agreement, and
therefore will lack any sort of privacy consideration (although I
admit what's currently in the PA is fairly benign).

> I was saying that short run the set of IdPs prepared to deal with the
> fallout from interfed probably matches the R&S tag pretty well so that's a
> reasonable way to do opt-in without deploying anything else.
>
> And I was saying that longer term the way to deal with IdPs that want to
> be left alone is to have them opt-out of discovery and be tagged for that.

The above pair of items make total sense to me. These could be added
to the TODO list.

> We can name the tag http://incommon.org/gretagarbo

:-)

Tom