I'm still fuddling with ways to send messages between actors and systems outside LV. I started using zeromq in LV, python, and node.js. I found it really easy to get them all talking to each other.
Would it make sense to create a zmq version of your messengers, and base actors on them?
I see this as a possible way to drive my existing event-based actor code from outside.
Sorry, I missed this post in February.
I have looked at zmq before, but have always been stopped by the paradigm mismatch between "people communicating by email", which is the analogy that Messenger Library follows, and "sockets on steroids" that zmq is.
As it says on zeromq.org , "ZeroMQ looks like an embeddable networking library but acts like a concurrency framework." ZeroMQ is a different concurrency framework than an "actor model" motivated one like "Messenger Library". ZMQ seems designed to build more specialist components that my "actors"; with the messaging patterns inherent to the sockets. It is not clear how I can interface these specialist sockets.