@ramkumarkb : If you mean that the facts are Java objects, I don’t see any barrier to that; JavaBeans are a very normal Clara fact. If you mean working in Clojure with a Java session of the sort in the docs you linked, you can interact with it via Clojure’s Java interop; see https://github.com/rbrush/clara-rules/blob/master/src/test/clojure/clara/test_java.clj#L42 for an example

@devn I think https://github.com/rbrush/clara-rules/blob/master/src/main/clojure/clara/tools/inspect.clj#L67 has the info you’d need for truth maintenance governed facts anyway, clara-tools builds on this https://github.com/rbrush/clara-tools/blob/master/src/main/clojure/clara/tools/impl/facts.clj

My understanding is that this ultimately works because Clara stores logical insertions in the activation-memory

I’m curious - do you have a particular sort of case in mind?

The graphs in clara-tools seem to work OK, though when I last poked around at them there seemed to be some oddities around accumulators, but I haven’t looked at it much yet

@wparker Our rules are pretty DAG-y, and continue to grow in number

for the sake of development, understanding the flow, it'd be nice to output the rules as a graph for a particular run

Ah, so more of a static thing than as a web app like clara-tools perhaps

I want to show someone who works directly with a customer the decision tree

though, as time goes on, a more interactive tool could be nice, to simulate what might happen if a rule were added to do X

yeah, something like a static doc visualization to hand someone could be useful