Two weeks ago, eWeek published an article on AMQ. From the article and the comment by James, AMQ is an attempt to standardize the ActiveMQ protocol. I heard some rumours about this a few months ago and think it’s an interesting idea. As Sean write, JMS is our current tool but it only specifies the API. What the implementation then does is beyond our reach and we therefore have to turn to integrating different messaging products. Getting messaging products to communicate between different implementations, possible even across enterprise borders could greatly simplify an integration world that today mainly consists of HTTP, FTP and EDI traffic. Having HTTP (REST) for synchronous, non-reliable traffic and a
asynchronous, reliable wide-spread protocol would be amazing.
I’m not all that convinced that the big players, particulary IBM, is all that interested in this work. IBM practically ows the messaging market with MQ and currently have three different messaging products (MQ, MQe, WAS 6 messaging) and protocols in the market. If IBM would be interested in turning all of this on its end, AMQ would need to be supported by one of the major standards bodies, but it sounds like it’s not. At least not yet.
IETF already got XMPP/Jabber that in many respects attempts to do the same thing, maybe that’s where AMQ should have started?
tags: Messaging/JMS