RE: [xacml] another small time/date issue

["Daniel Engovatov" <dengovatov@bea.com>]

>Based on this, I don't think a clarification about the current
date/time
>attributes has any effect on any other attributes in the system. All
I'm
>worried about is that since there's no language explaining how current
>values must be provided, it leaves different implementations open to
>different models.

I would think that such specification would be out of the scope of the
XACML standard.   How would you realistically specify that?  Establish
accuracy?
How do you deal with geographically distributed systems, etc?  

We are quickly going down the path of making the authorization request
and language standard into some sort of information management system,
which, I would think is hardly possible.

We say that "current-time" and friends should be provided and expected
to be defined, why is not it sufficient?  We can not adjust all
implementation clocks anyway.

Daniel.