Shibboleth Developers

[Halm Reusser <>]

On 15.11.10 19:31, Chad La Joie wrote:
> So for me, at this point, I am mostly settled on the model where we
> leave it in but have it off by default. What I really want to ensure
>  though is that our APIs are developed such that they are easy to
> extend or chopped up and reused for people who want to do drastically
> different UIs. I feel pretty confident in saying that outside the
> very simple uApprove/card selector UI we have today, we have no idea
> what UIs people really find usable.

Chad, I really appreciate your pragmatic way of thinking.
Most probably is that the solution which fits for the moment - let's see
how it will be used in the future and which kind of complaints drop in ;-)


On 15.11.10 18:47, Peter Williams wrote:
> Perhaps play with the openid consent capability at myopenid.com.
> It's pretty simple (and is expressed in the notions of the UCI-class
> protocol itself, to some degree). Its worth looking at what others
> have done, in a compare and contrast exercise.

Yes, that would really worthwhile in the future - to have a look at
those solutions. Thanks for the example. May be the support of
"different identities" is not as much asked in the "Shib World" - just
my impression...

On 15.11.10 19:10, Tom Scavo wrote:
> Are you suggesting that global consent be applied on an
> attribute-by-attribute basis? That doesn't seem like the correct
> granularity. Have you looked at the Cardspace metaphor recently? (I
> haven't, but I wonder if it applies here.)

I haven't looked at the Cardspace metaphor. But I'll catch it up. Thanks.
Yes, my suggestion was indeed of this granularity.

-Halm