RATIONALE

"Stefik, MOLGEN"

M. Stefik, Planning with Constraints (MOLGEN: Part 1). Artificial
Intelligence, Vol. 16, No. 2, North-Holland, 1981, pp. 111-140 

M. Stefik, Planning and Meta-planning(MOLGEN: Part 2). Artificial
Intelligence, Vol. 16, No. 2, North-Holland, 1981, pp. 141-169  



MOLGEN's domain is the "design" of genetics experiments --- at the very
least that raises the issue of what the difference is between planning
and design!  There were two MOLGEN systems, one based on "skeletal
plans" (i.e., outlines of plans that got filled it based on the given
problem) and the one we'll study that constructs plans from basic lab
actions.  

It is worth including because of the use of Means Ends Analysis to
drive the planning/design process, the clever use of constraint
posting, the multi-level architecture, and the general "least
commitment" approach.  MOLGEN uses a significant amount of knowledge
about lab operations and the items being operated on in order to
reason out what constraints must need to be true on the inputs to an
operation so that the outputs are constrained appropriately.

If this is a planning system, can the same approach be used for
design, and vice versa?