Open Issues

• Mathematica infix notation.
• Are some classes non-terminal, abstract classes that can't be directly instantiated? Something along these lines came up in one of the discussions.
• Is there any real advantage of using same and different instead of = and \= with same restriction to the :conditions field of definitions when applied to non-quantity-values?
• Define properties of relations such as: associative, transitive, commutative.
• Currently continuous-quantities are defined. Do we want to allow for piecewise continuous quantities as well?
• Atomic facts in the domain theory outside of the scenario definitions and rules. They can be asserted in the consequences of MF and entity definitions.
• nth derivatives, compose operators (?)
• interclass relationships (exclusivity of siblings)
• The relation exists is not being used. Should it be allowed in defRule or defModelFragment? Consensus seems to be that it should not be in the user language, and be rather carefully used internally. At them moment it is used by defModelFragment when there are no superclasses. It is being implicitly used via defined for attributes and quantities.
• We clearly need to be able to extend the notion of (scalar) quantities to include vectors as well. Currently, we don't have much experience working with such objects. Adam suggests that we extend the language to cover them after individuals have accrued experience with appropriate syntax and semantics.
• Tables (e.g., viscosity is defined via data tables on temperature, etc).
• Non-Numeric (discontinuous) events.
• Procedural specifications (e.g., NA routines, problem solving operations, such as ``do this operation, then this operation...'').
• default initial values (non-defaults can be done via defScenario)