What is in the future for JADE?

Question

I'm starting my thesis on Agents and Smart Objects interaction and I'd like to know what's in the future for JADE, the Java Agent framework. I find the whole concept of agents, programmable behaviors, federations and their help in solving Artificial Intelligence problems very interesting but will it always be an academic field, like Haskell? What's being done with JADE?

Answer 1

I created a system for my PhD - using JADE as the framework - where agents played music with human musicians. There's a bit of a write up (and a link to the thesis) at http://www.mo-seph.com/academic/musicalagents

In the end, I didn't use a lot of JADE, and found it quite heavyweight for the kind of work I was doing. The communications library was useful (if a bit hard work) and the message queuing side of it worked OK. However, I quickly found that most of the behaviour I wanted to implement needed to be so heavily customised that the JADE framework only really supported the transfer of information.

I think the usefulness of the framework depends hugely on what the question is. It seems that a lot of the usefulness of JADE was around it's links to the FIPA agent communication languages (http://www.fipa.org/) and the communications infrastructure in general. If that's what you're interested in, then it might be quite useful. If you're more concerned with the intelligence in individual agents, it might be a useful framework, but it won't help much with the design of the intelligence.