TLDR

Will Richardson mentions this recently.

I have been thinking a lot about how learning objects like CLIPS can be precise and concise at the same time.

Did you get through all that?