Computed ID Data Connector

[Tom Scavo]

On Fri, Mar 28, 2014 at 10:05 AM, Cantor, Scott <cantor.2 at osu.edu> wrote:
> On 3/28/14, 9:41 AM, "Tom Scavo" <trscavo at gmail.com> wrote:
>
>>On Fri, Mar 28, 2014 at 9:05 AM, Rod Widdowson <rdw at steadingsoftware.com>
>>wrote:
>>>> 1. Does the IdP entityID enter into the actual computation?
>>>
>>> Yes
>>
>>I was afraid of that. May I ask why that is and/or why that has to be?
>
> It was probably done to avoid conflicts in the case of a multi-homed IdP,
> but that's just me speculating.

I'm not sure why that matters. The same computed ID paired with a
different IdP entityID is in fact a different Persistent NameID,
right?

> Creating a new connector as a copy is quite trivial so it doesn't really
> matter that much what any specific one does if there's a particular
> deployment need to do something different, unless we're talking about
> existing values already propagated.

Right, but I wanted to ask these questions first before I ran off and
cooked up something that turns out to be flawed.

> You also have to remember that until very recently, that was essentially
> viewed as "do not use" code, so I doubt a lot of attention was paid to its
> design in V2 when it was ported.

Understood.

Tom