Is this correct behaviour ?
Yes, the way you do it, it is.
If it is ? Couldn't this be done better ?
Yes, by using swimlanes
Thanks for your response.
I made the change to swimlanes; but it doesn't seem to affect the behavior of creating records in the JBPM_POOLEDACTOR table for every task instance.
I guess this is just how it works. Perhaps I'll repost on the JBPM-DEV list for comment.
huh? do you have a unit test that show this? Might be that the actual pooled actor of the swimlane is copied runtime, but that is not what you want. Suppose you change the assignment in the swimlane... then all pooled actors of taskinstances pointing to that swimlane would have the wrong actors.... hmmm...