On Tue, Jul 10, 2001 at 11:12:56AM -0600, Hilarie Orman wrote:
> The 'explicitly addressed' was added to explicitly address your
> concerns that the IP addressing model might be violated.

parsing... parsing...

I don't *think* I've raised any such concerns... my primary concern
has been the violation of the implied trust model on the Web, not
IP's addressing model.


> Now you appear to be getting at application level addressing or
> 'targeting' (what do mean by 'arbitrary content'?).

The phrase 'arbitary message content' was attempting to communicate
the idea that some part of a message passing through an intermediary
nominiates that message for modification, etc., rather than some
out-of-band ruleset, etc.


> The security model can encompass delegation within the message to
> any degree of specificity, but it seems to me that classes of
> delegation are the most useful principals - 'proxy', 'surrogate',
> 'edge proxy'.

In the infamous words of Pauline Hanson, "Please explain?"


> For most publishers, these delegations, if required, likely will be
> added by reverse proxies that modify content based on pattern
> matching.

This statement seems to make a number of assumptions about the
publication processes used, the nature of the content, the
publisher's relationship with the 'reverse proxies' (n.b. - WREC
discouraged this term in favour of surrogates), and the nature of the
services which will be applied.



> On Tue, Jul 10, 2001 at 09:39:17AM -0700, Michael W. Condry wrote:
> > At 09:09 AM 7/10/2001 -0700, Mark Nottingham wrote:
> > 
> > 
> > >SOAP intermediaries must be explicitly targetted by the message
> > >(using the 'actor' attribute). In this respect, they are completely
> > >unlike the OPES model.
> > 
> > The charter says the intermediaries are explicitly addressed. I am
> > a bit confused by your comment.
> 
> Does the charter state that they will be targetted by an in-message
> mechanism (e.g., HTTP response header, arbitrary message content,
> etc.)?
> 
> 
> -- 
> Mark Nottingham, Research Scientist
> Akamai Technologies (San Mateo, CA USA)
> 

-- 
Mark Nottingham, Research Scientist
Akamai Technologies (San Mateo, CA USA)

Reply via email to