Re: [wss] Re: [wss-comment] recursive Security Token References

[Anthony Nadalin <drsecure@us.ibm.com>]

That's correct Ron that the usage is a per reference instance, but since some tokens have closed content model there may be a need to associate a usage with them and since we have no general SecurityToken one uses a STR with a usage attribute on a token, the usage of wsse:embedded allows any token otherwise there is no element to contain a token that does not fit the type define in WSS.

Anthony Nadalin | Work 512.838.0085 | Cell 512.289.4122
Ron Monzillo <Ronald.Monzillo@Sun.COM>

Ron Monzillo <Ronald.Monzillo@Sun.COM>

09/01/2005 12:33 PM

Please respond to
Ronald.Monzillo

To	Anthony Nadalin/Austin/IBM@IBMUS
cc	"Conor P. Cahill" <concahill@aol.com>, Tech Rams <techmailing@yahoo.com>, "DeMartini, Thomas" <Thomas.DeMartini@CONTENTGUARD.COM>, wss@lists.oasis-open.org, wss-comment@lists.oasis-open.org
Subject	Re: [wss] Re: [wss-comment] recursive Security Token References

Anthony Nadalin wrote:
> Ron, I don't see that inconsistency, we went through these usecases for
> embedded, and there are cases where you want to be able to put a usage
> attribute on a token and include that in-line as you need a way to then
> reference that in that usage context.

any usage attributes pertain to the reference not the token; independent
of whether or not the token is embedded.

How does a need yo hang usage attributes on an str, create a need for
the embedded str?