[WG-IdP-Selection] Kantara IDP Selection Hillsboro meeting March 10th

[philippe...@orange-ftgroup.com]

Dear all,

 

Thanks to all people who attended this F2F plenary in Hillsboro and made that meeting so efficient !

 

Please find below a link to the minutes of our two sessions, wednesday March 10th afternoon

http://kantarainitiative.org/confluence/display/WGIDPSEL/IdP+Selection+WG+Minutes+-+2010-03-10

You will find inside all needed links to the docs that were worked during the session.

 

Next Call

As our next call was initially scheduled tomorrow tuesday march 16th, I suggest to cancel it to allow action points to be completed for the next time, unless anyone objects.

Our Next Call should then take place on March 30th, as you will see inside the minutes.

 

Once again, a great "thanks" to you...

 

Kind regards,

Philippe

 

*********************************
This message and any attachments (the "message") are confidential and intended solely for the addressees. 
Any unauthorised use or dissemination is prohibited.
Messages are susceptible to alteration. 
France Telecom Group shall not be liable for the message if altered, changed or falsified.
If you are not the intended addressee of this message, please cancel it immediately and inform the sender.
********************************

[Scott Cantor]

A couple of notes from reading the slides and proposed charter text (sorry I
couldn't call in, just too much conflicting):

Regarding the proposal about discovery related information in SAML metadata,
my project has been developing a proposal for a new metadata extension that
addresses things like display names and logos, and it should be making it's
way to OASIS at some point soon, not sure about our schedule yet.

We have elected not to go forward with the use of the Organization element
because there are existing practices around that element that conflict with
the requirements of discovery.

(Obviously others can still take proposals to OASIS on this, I'm just noting
our position for the record.)

The other minor comment I had was that the charter proposal (as the original
ULX charter does) equates OpenID and OAuth with SAML as formal standards,
but that's not the case. They're at most de facto, while SAML and IMI are
formal standards. It may seem a minor point, but accuracy matters in these
conversations when one is constantly fighting technical misinformation
already.

-- Scott

_______________________________________________
WG-IdP-Selection mailing list
WG-IdP-S...@kantarainitiative.org
http://kantarainitiative.org/mailman/listinfo/wg-idp-selection

[gael.go...@orange-ftgroup.com]

Thanks Scott for your comments on the slides and the charter.
Regarding your proposal for a new SAML metadata extension, is it something you can already share with us ? Is it different from the "SAML Metadata Extension for Entity Attributes" ?
As mentioned in the slides, one missing element might be also a way to express in the SAML Metadata the list of AuthnContext classes supported by an IDP (in the same way as the LoA status). Am I right or is it something that already exists in SAML specs ?
Concerning OpenID/XRDS, I will solicit Peter Davis in order to make progress on that topic too.
Thanks,
Gaël

-----Message d'origine-----
De : wg-idp-selec...@kantarainitiative.org [mailto:wg-idp-selec...@kantarainitiative.org] De la part de Scott Cantor
Envoyé : lundi 15 mars 2010 19:47
À : CLEMENT Philippe NAC AAD; wg-idp-s...@kantarainitiative.org
Objet : Re: [WG-IdP-Selection] Kantara IDP Selection Hillsboro meetingMarch 10th

[Scott Cantor]

> Regarding your proposal for a new SAML metadata extension, is it something
> you can already share with us?

Yes, but with several caveats. I didn't write the material, not all of it is
agreed to even internally yet, and it definitely isn't ready to submit to
OASIS yet.

https://spaces.internet2.edu/display/~laj...@idp.protectnetwork.org/DSUI
https://spaces.internet2.edu/download/attachments/9731/saml_ds_login_ui_02.o
dt

This is just work in progress material.

> Is it different from the "SAML Metadata
> Extension for Entity Attributes" ?

Yes, it has no connection. Using that extension results in a lot of bloat
without any compelling benefits, so that isn't the current proposal. One
problem is that some of the data for discovery is not just simple strings,
and most products that deal with SAML attributes (such as would be in that
original extension) can't be bothered to support non-trivial attribute
values.

> As mentioned in the slides, one missing element might be also a way to
> express in the SAML Metadata the list of AuthnContext classes supported by
> an IDP (in the same way as the LoA status). Am I right or is it something
> that already exists in SAML specs ?

No, it hasn't been done, but an attribute could be defined for that and the
EntityAttributes extension used, I suspect, since we're doing that for LOA
anyway.