> "Participation" includes "authorization" and is part of the security
> model.
If you are saying that explicit authorization is a necessary component to
participation, that sounds okay. If you are saying that authorization is
implicit in participation, I'd have a big problem with that.
> The notions of integrity and authorization are obligatory
> in the WG documents.
There is no WG yet, so there are no WG documents. Anything written prior
to the charter being approved is subject to change, or even to being
discarded. We should not assume that the WG will adopt it once formed.
> Other WG's are concentrating on the network and transport layer
> intermediaries. We've got HTTP and possibly RTP.
>
> We have no control over whether or not functional layer
> separation is abandoned.
True, we don't have control over it, but that doesn't mean it's a good
idea for a WG group to standardize a way of breaking layer separation.
And there are lots of folks who keep trying to do that. I have no way of
knowing whether whether they're represented in this proposal or not - it's
too vague to tell.
> The detailed wording of the charter comes from Area Directors;
> I don't like the connotations of "arbitrary" and can see no reason
> for using it. However, it is scary to imagine the depth of the
> paranoia that jumps from there to the end of the Internet
> architecture.
One person's paranoia is another's experience. Surely you will admit that
firewalls, NATs and interception proxies exist, that they do harm
interoperability, and that many of them act without the consent of their users
and routinely change traffic payloads on the fly?
> The architecture is motivated by notions of dynamic content
> and the notion of distributed semantic evaluation. It is a logical
> general of caching.
It would be useful to say something about this in the charter.
> The technical issue of whether or not the IP addresses are honored
> is outside the scope of the charter and the technology developed
> therein.
Perhaps they are indeed orthogonal. But again these things do exist,
and the charter can easily be read to say that the WG will be working
on these things. There are many kinds of "intermediaries" in the
Internet, and saying that they can do "arbitrary" things is leaving
a lot of wiggle room.
I also think that even extending the notion of intermediaries within HTTP
is operationally dubious - intermediaries have been HTTP's biggest problem
to date. But in relation to the scope of the current proposed charter,
that's a miniscule detail.
Keith