I am confused in a way. I would have thought that all of the token types are defined on a per mechanism method, thus re-using the same value would have no theoretical problems. I recognize that they could present an issue if you use a single code base for implementing multiple mechanisms and use a single switch statement to deal with the token types.
I guess my question would be why would we want to look at the 4121 token type registry for anything given that it should be mechanism specific? Jim > -----Original Message----- > From: Sam Hartman [mailto:[email protected]] > Sent: Sunday, October 09, 2011 8:06 AM > To: Jim Schaad > Cc: [email protected] > Subject: Re: Trying to build the first GSS-API token ---- draft-ietf-abfab-gss- > eap-02 > > I will attempt to standardize on subtoken. > I definitely agree we need more clarity around the RFC 4121 token type > registry. > We use a different token type for our context tokens from Kerberos because > they have nothing in common. _______________________________________________ abfab mailing list [email protected] https://www.ietf.org/mailman/listinfo/abfab
