Re: [HOKEY] Last Call: draft-ietf-hokey-reauth-ps

Charles Clancy wrote:
> It was decided to split up the documentation into the "problem" and the
> "solution".  The PS document described the reauth problem and gave
> design goals for a fast reauth protocol.  It motivates WHY we need fast
> reauth, but not HOW to implement it (other than high-level design
> goals).  The "solutions" documents then provided protocols that met
> those design goals, justifying their approach.

  The problem statement can contain a description of the current
architecture.  That description can refer to *existing* systems without
proposing a solution to the hokey problem.

  To put it another way, I haven't followed hokey since the beginning.
There appears to be knowledge captured on the list that isn't available
in the documents.

> Addressing AAA proxies will occur in detail in draft-ietf-hokey-key-mgm
> (text not there yet).

  Again... how do AAA proxies fit into the problem statement?  They are
first mentioned in Section 5.4.  The term "AAA proxy" is not defined in
Section 2, and there is no reference in that section to any document
that defines the term.  The document just assumes that everyone knows
what an AAA proxy is, and how it fits into the problem statement.

  I will try to come up with suggested text to address my comments.

  Alan DeKok.
_______________________________________________
HOKEY mailing list
HOKEY@ietf.org
http://www.ietf.org/mailman/listinfo/hokey