ietf-openproxy
[Top] [All Lists]

RE: WG Review: Open Pluggable Edge Services (opes)

2001-06-20 19:07:33

I think you could do most of this with signatures.  You can make sure
that if someone does modify the original, you know about it, and that
means you can make sure only someone you've authorized can modify the
material.  However in the case where an intermediary is supposed to
replace material for you (e.g. change encoding for a low bandwidth
device), you can't be sure that what eventually arrives has any relation
to what was originally sent.

...Scott

On 20 Jun 2001 at 13:41 -0700, James Binder apparently wrote:
In essence, I think you would need two channels (bound together somehow) --
One with non-mutable content (i.e., what the user is buying from the origin
server) and mutable content (local site specific ads or what not).

/jsb

-----Original Message-----
From: sbrim(_at_)cisco(_dot_)com [mailto:sbrim(_at_)cisco(_dot_)com]
Sent: Wednesday, June 20, 2001 12:53 PM
To: Abbie Barbir
Cc: ietf(_at_)ietf(_dot_)org; ietf-openproxy(_at_)imc(_dot_)org
Subject: RE: WG Review: Open Pluggable Edge Services (opes)

On 20 Jun 2001 at 15:45 -0400, Abbie Barbir apparently wrote:
1. Ensure that both end points are involved in the decision process.
2. Ensure that OPES's type devices does not allow unauthorized
interception
   of traffic between source and destination.

How would you do either of these?  If the OPES box is allowed to do its
work at all, I don't think either of these is enforceable.