RE: Thoughts about path and intermediaries

>So in summary, the notion of path that interests me at the 
>moment, if any, 
>is at the abstract level of headers and actors.  In any case, I would 
>welcome discussion of my presumption that actors are, at least in the 
>general case, named at a level that is potentially independent of 
>particular bindings or physical end points.  Of course, nothing would 
>prevent one from basing an actor name on a binding-specific URI in 
>situations where that is in fact a sensible strategy.  Comments?

I agree that either are possible (and allowed by current SOAP as a side
note). In a sense all URIs are abstract names but the question still remains
of how late one would want to bind a name in any given XML Protocol message
instance. One way to describe this might be to say that both an XML Protocol
sender and an XML Protocol receiver may provide a binding for any given URI
including an actor URI. 

For the case of the receiver providing the binding I think there are two
sub-cases that we have to consider:

1) A sender wants to address a "class of service" like for example a cache.
The sender wants to be able to control the cache but may not know exactly
where it is in the path. A recipient does the late binding to its specific
cache service identified by the actor URI.

2) The sender knows exactly which recipient it wants to address but there is
no 1:1 mapping between the actor URI and the URI used for indicating the
destination. 

In 2) where the sender knows which recipient it wants to address then I
think there are two questions that come to mind:

a) How does a sender know which endpoints/actors are exported by a given
receiver and the receiver will in fact do the late binding? 

b) How can a sender trust the information in 1)?

SOAP provides the special case of "the next guy" because the trust model for
this case is simple but in the general case, the trust model doesn't seem to
be that simple and might have to involve DSIG.

One can of course ask the exact same questions in case there *is* a 1:1
mapping but there we at least have some notion of trusting DNS although it
might be hard to do these days.

Henrik

Received on Thursday, 15 February 2001 23:00:41 UTC