Surrogate Deployment
====================

- So far as the container is concerned, a Surrogate is just another kind of
application that is handled by a deployer.
- The surrogate deployer sets up on an interface of some kind looking for 
surrogates to deploy.  As per the surrogate over IP spec, this will involve
setting up a multicast listener that responds to requests on the surrogate
interface.
- One thing to note - the surrogate deployer might be listening to a  different
interface than the Jini workgroup is running on.  In the case of a non-TCP/IP
surrogate, this is guaranteed to be the case.
- For instance, let's say we have a PBX or asterisk implementation that wants
to publish a Jini service by means of a surrogate.  The scenario is that the 
surrogate talks to the PBX by means of UDP or simple TCP sockets.  If the PBX
was unable to support a secured protocol, then a node might 
have a connection on a private network to the PBX, but then might have a normal
network connection for other clients of the PBX surrogate.  Since the surrogate 
runs in the Java/Jini environment, it can use the full JERI security stack.