see inline,
-----Original Message-----
From: Graham Klyne [mailto:GK-lists(_at_)ninebynine(_dot_)org]
Sent: Tuesday, May 21, 2002 3:21 PM
To: Barbir, Abbie [CAR:1A00:EXCH]
Cc: OPES Group
Subject: RE: WG Last Call: draft-ietf-opes-architecture-00
SNIP
The objective here is to allow a qualified administrator to
trace what
happended to the data when the end user gets errors.
In which case, I don't think that's an adequate level of tracing
capability. That is, I don't think it fully addresses the
spirit of the
IAB requirements.
I refer you to the example I mentioned, of a problem with an ISP's
transparent caching. I don't think it's acceptable that only
the ISP's
administrator can find out what happened, because in practice
the ISP won't
bother and will spin some lie about it being (say) a browser problem.
I think it's crucial to put diagnostic capability in the hands of the
endpoint on whose behalf (authority) the service is being
performed. (Even
if the endpoint isn't given any choice about having the service.)
#g
-------------------
Graham Klyne
<GK(_at_)NineByNine(_dot_)org>
This is why header extensions may be needed. simply keeping a log will
suffer from the same problems that u mention plus it does not scale.
abbie