Then divide these entities into the environment and the system. Superimpose this division on to the UML-class-model-expressed domain model. You can do this superimposition by drawing a shape enclosing the environment entities and another shape enclosing the system entities; then label each shape as ENV or SYS as the case may be, as on page 21 of the slides for Classes & Concepts. Alternatively, you could just mark each domain model entity with ``ENV'', ``SYS'', or both, as the case may be, as on page 47 of the same slides.
If you have done this superimposition correctly, you should be able to see clearly which entities are in the interface.
The hardest part of this exericise is coming up with a concise, consistent, and complete set of use cases that allow the user to do what he or she needs to without having too many different ways of doing any one thing.
Describe each feature as a use case, i.e., give a simple imperative sentence as if when the user says the imperative sentence as a command, Matchmaker will obey and do the use case.
You do not have to develop any scenarios yet. However, your imperative sentences must be complete enough that the customer can understand what the feature does. Thus, a phrase, such as ``the questionnaire'', does not cut the mustard because it does not explain what happens with the questionnaire.
Build a use case diagram that shows the system boundary, all actors, all use cases and that associates each use case with each actor that may do it. Use the notation of the use case diagram on page 22 of the slides for Classes & Concepts.
The two diagrams should be consistent in that