SPA Conference session: Behaviour Modelling -- Muddles and Opportunities | |||
| One-line description: | An exploration of concepts and techniques in behaviour modelling and their implications | ||
| Session format: | Tutorial (150 mins) [read about the different session types] | ||
| Abstract: | Anyone who has been involved in business system development will agree that the behaviour of the software is key. Behaviour defines how the system reacts to its environment (users and other systems) and, much more than the structure of the software, is what interests the users and determines the systems role and function in their world. But there is a malaise in the way we understand and specify behaviour, evidenced by: - Limited use of models of software behaviour compared to models of software structure (Class Diagrams and Component Models) - Complex behavioural model semantics (UML2 statechart) - Continuing notation wars (UML Activity Diagrams vs. BPMN vs. BPEL vs. Petri Nets) This session will explore and explain thinking in the area of behaviour modelling and its implications for the way we specify and develop software. The main topics will be: - What do we mean by behaviour? What is important from the requirements point of view? - Key concepts in behaviour specification: Contracts, Constraints and Business Rules - UML behavioural notations: a bit of a semantic muddle - Aligning semantics to concepts: clearing the semantic muddle - Implications for ideas of inheritance and re-use - Implications for model execution and the system development lifecycle. | ||
| Audience background: | An interest in modelling and such questions as: - Can we execute models? If so, what sort of models? - Can behaviour be inherited (as attributes and methods can)? - How does behaviour specification fit into the software development process? Some familiarity with: - the behaviour modelling notations of UML (Activity Diagrams, State Charts) - the ideas of Design By Contract (Pre-conditions and Post-conditions) - MDA, DSLs, code generation and such things. | ||
| Benefits of participating: | Insights into the concepts used to specify behaviour and what they mean. An understanding of the strengths and limitations of UML in the behaviour modelling area. The shared delight of some polemical digs at UML. Some good discussions (some of the material is likely to be controversial). | ||
| Materials provided: | - Presentation material - Some exercises for people to do. These will take the form short narrative specifications which participants have to analyse from the behaviour point of view, and build some simple behaviour models. | ||
| Process: | Lecturing, excercises and discussion sessions on the exercises. Participants will work in groups on the exercises. | ||
| Detailed timetable: | |||
| Outputs: | Lecture material plus notes from the exercise discussions. | ||
| History: | None. I have practised and published extensively on this subject, and presented parts of this material at various conferences and meetings. But I have never "put it all together" in this way. | ||
| Presenters | |||
| 1. Ashley McNeile Metamaxim Ltd. |
2. | 3. | |