An open discussion list for topics related to the eduGAIN interfederation service.

[Tom Scavo <trscavo AT internet2.edu>]

On Wed, Aug 10, 2016 at 3:10 AM, Lukas Hämmerle
<lukas.haemmerle AT switch.ch> wrote:
>
> On 09.08.16 20:35, Tom Scavo wrote:
>
>> - The phrase "Organisations/Identity Providers" is not relevant to
>> InCommon since orgs are allowed to have multiple IdPs. In any case,
>> the two are not equivalent. Historically, conflating the two has led
>> to widespread misuse of md:OrganizationDisplayName.
>
> So far I have been only aware of the opposite situation where multiple
> organisations have one IdP (is the case in some H&S federations).
> Therefore, the organisations-IdP relationship is a tricky one in
> general.

Indeed. The history of md:OrganizationDisplayName confirms that.

> Do you have an idea of how many organisations in InCommon
> operate several IdPs and how many such organisations operate?

Yes, this is documented on our (dynamic) org info page:

https://incommon.org/federation/info/all-orgs.html

> I'm just
> wondering why an organisation would invest the effort to operate
> multiple IdPs (unless one is for test and one is for production)

I totally agree but that is a common use case worldwide. You can
browse this (dynamic) list of IdP DisplayNames to get a feel for how
many test IdPs there are in metadata:
https://spaces.internet2.edu/x/2IDmBQ

But I think I've (unintentionally) lured you into a rathole. All I'm
really asking for is:

s^Organisations/Identity Providers^Identity Providers^

Do you agree?

>> - Can you add a row of totals at the bottom?
>
> What would you be most interested in? Just calculating the total would
> be easy but in case you are interested in the total number of (unique)
> IdPs in production federations, this would not be the total of the above
> columns (some IdPs probably are in multiple federations).

Yes, I'm sure that's true but there can't be many such IdPs, right?
(that's a rhetorical question) In any case, I think a total would be
helpful. It gives an estimate of "all the IdPs in the world."

Thanks,

Tom