On Fri, Dec 9, 2011 at 1:34 PM, Jim Schaad <[email protected]> wrote:
> I am not saying that we need to say what the transport is, but if you have a 
> neophyte looking at the document and trying to figure out what is happening 
> they are going to start assuming that GSS-API apparently has a transport as 
> part of it.  As we know this is incorrect.

It's not incorrect though: GSS per-message tokens do in fact form a
secure channel/transport.  That's often (but not always) not as
convenient as TLS, and for some mechanisms (e.g., ones based on bearer
tokens), not secure.

> Additionally we may want to specify that the transport has some properties - 
> such as channel binding - that may or may not be of interest.  What are the 
> issues of using a non-secure vs a secure transport and so forth.

Sure.  I don't mind some text to this effect in the architecture
document, but it feels a lot like an update to RFC2743.

Nico
--
_______________________________________________
abfab mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/abfab

Reply via email to