@Jan The Drools team always appreciates feedback, so I would like to ask you to clarify a few of your statements if that is ok?
What gives you the impression that Drools Flow does not perform well in the Java world? An action node can contain any Java code you want. You have direct access to your process variables (as if they were local variables) and you have clear access to your context through a kcontext object. You can easily define processes in Java (if you don't like XML) using a fluent API. The engine itself is a simple Java POJO component that can run embedded (without any db even). What features are we missing to make it even more Java-friendly? Let's suggest you have an application where you're not that interested in rules. Why do you think Drools Flow has a more complex architecture in that case (basically both engines are implemented as a simple state machine, if you don't use rules the underlying technology is very similar)? Does the fact that it is also possible to use rules make the overall solution more complex? We tried to keep the APIs separate as much as possible though (check out http://anonsvn.jboss.org/repos/labs/labs/jbossrules/trunk/drools-api/src/main/java/org/drools/runtime/process/ProcessRuntime.java, I don't think it can get much easier than that?). View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4254504#4254504 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4254504 _______________________________________________ jboss-user mailing list [email protected] https://lists.jboss.org/mailman/listinfo/jboss-user
