Joined: 09/09/2013 04:03:02
The original P2P functionality did not implement direct client-to-client communication; all communication using the P2P API took place via the server (i.e. client -> server -> serviceprocess -> server -> client).
With the current API, this is analogous to, for example, using a queue to hold client messages, and to have a second process consuming these messages and publishing appropriate responses to the client, either via another queue, or via a channel or a datagroup. An example of such a system can be found at:
Using the old P2P API, the "serviceprocess" mentioned above was typically implemented using one of the enterprise APIs (e.g. Java), and would correspond to the "Responder" in the above example.
Apologies for the confusion caused by old documentation still existing on the current website; we will get this tidied up immediately!