RE: [higgins-dev] Items for F2F agenda

["Tom Doman" <TDoman@xxxxxxxxxx>]

Mary,

I discussed these with Jim and they all sound like great topics to me.  Although I don't know if these are topics of the highest level, if you know what I mean.  These are topics that are very germane to the implementation of Higgins components but may not be of general interest.  That said, I'm happy to start these at a high level and discuss each for, say 15 minutes.  I don't know if you have the flexibility, but I'd like to leave you all the ability the assess the follow on applicability of these topics.  How does that sound?

I think I'd like to start with the less complicated and\or more immediate stuff first.  Say, 3, 1, 2, in that order.  I'll try to put together a tighter outline before we start but I'm going to be getting ready to travel and traveling most of the next 24 hours and I don't have a long lived battery.

Tom

>>> "Mary Ruddy" <mary@xxxxxxxxxxxxxxxxx> 11/14/06 3:44 PM >>>
Have added a placeholder for focused IdAS topics.  Please go ahead and
propose a tighter outline with timeframes.

-----Original Message-----
From: higgins-dev-bounces@xxxxxxxxxxx 
[mailto:higgins-dev-bounces@xxxxxxxxxxx] On Behalf Of Jim Sermersheim
Sent: Tuesday, November 14, 2006 5:02 PM
To: higgins-dev@xxxxxxxxxxx 
Subject: [higgins-dev] Items for F2F agenda


Can we get these IdAS items added?
 
1) Alternate schema formats:  Valery checked in some initial stuff -- it
would be good to review that and see if it meets everyone's needs.  One
thing we've been thinking of adding to the schema is the association
between attribute types and filter comparators, and between filter
comparators and assertion types
 
2) Further refinement of update operations.  We keep pushing off
atomicity rules / transaction semantics.  Also there are some other
recent things that have come up.
 
3) The need for common configuration and policy for contexts.  The
sample code I sent illustrates some stuff we're doing here, but it's
completely context-specific.  It would be nice if an IdAS deployer
didn't have to learn a new config format for every CP being deployed.
Also in terms of policy -- how does the IdAS consumer discover what
kind(s) of AuthN identities may be passed to IContext.open?
 
Tom and I will discuss these further and maybe propose a tighter outline
with proposed timeframes.
 
Jim