RE: [security-services] A Modest Proposal about Attribute Names

["Eve L. Maler" <eve.maler@sun.com>]

At 08:48 AM 2/12/02 -0800, Hallam-Baker, Phillip wrote:

>Pleeeeeeaaassseee!
>
>We are going arround in circles here. The single URI attribute would bring
>us almost full circle back to the original proposal in XTASS.
>
>The argument over whether we have one attribute URI or the current structure
>has been discussed at length at a face to face.
>
>
>At present the spec only allows one name per namespace.

Where does it say that?  I just checked core-26, and I don't see any such 
restriction.  If it did, and if this were something the group had agreed 
on, then I would agree that there's no point to having both the name and 
the namespace, since the rule would force a different "namespace" for each 
name.  (The word "namespace" would then be inappropriate as well as 
confusing; AttributeNameIdentifier would be better.)

>Separating the two
>ideas would have more relevance if an attribute statement could have
>multiple attributes with different names in the same namespace that is
>declared only once.
>
>Ie.
>
>namespace = http:/fred.org/fredness
>         name = Alpha
>         name = Beta
>
>As it is the two can be combined quite easily:
>
>http:/fred.org/fredness#Alpha
>
>This would indicate some modest schema changes and eliminate one of the
>empty elements that Eve hateth so much yea even unto the fifth generation,
>etc.

They can be combined, but if you require them to be combined, it's not a 
namespace but a unique attribute name identifier.


>The most obvious way to represent X.500 attributes would be to use the OID
>URN:
>
>namespace=OID:12.3.42.3.42.1.1.1
>
>There is no need for a name in this instance.
>
>I will do some research to find out the current status of the OID URI, if
>the toplevel scheme did not go through the URN scheme must have.

         Eve
--
Eve Maler                                    +1 781 442 3190
Sun Microsystems XML Technology Center   eve.maler @ sun.com