ietf-openproxy
[Top] [All Lists]

RE: Chaining of Callout Servers

2002-07-25 16:55:20

From: The Purple Streak, Hilarie Orman 
[mailto:ho(_at_)alum(_dot_)mit(_dot_)edu]
[snip]
Could you give us an idea of what this complexity is?  I believe that
it is only the "this is a chained request/response" header. 

Since I was talking about managing the transitivity of the CSC caller's
trust, I'm keeping the context to such.  (Factoring in security aspects
will exacerbate the problem, ala complexity).

In the case of no chaining, the protocol only has to deal with
 O->A, a simple case.  O trusts A and has direct contact with A;
and that's end of it.

In the case of chaining, we have:
 O->A|B|C 
O only has direct contact with A; but has to decide whether to 
trust B or C, and to be reasonably assured that the pipeline only
comprises A, B, and C.  Something will have to be there to help
O make the decision, and such something will naturally add
complexity.

I might be wrong, but at least I'm being specific.

You wrong?  Well, if you were wrong, I don't know what wrong or
which wrong you're talking about.

That said, in light that during the Yokohama meeting it was
mentioned OPES was four months behind its deliverables,

OPES was chartered with its deliverables due immediately; this appears
to have been a deliberate and unexplained IESG decision.  No need to
think that the WG is doing something wrong by actually taking the
time to write the documents and discuss them, no matter what the
charter says.

Agreed, which was why I only went so far as sharing the sentiment
(of being schedule conscious), but not pushing it.

Joe Hui
Exodus, a Cable & Wireless service
========================================================



Hilarie


<Prev in Thread] Current Thread [Next in Thread>