Maui, HI, USA
Jan. 6, 2001 to Jan. 6, 2001
The Unified Modeling Language (UML) represents a significant milestone in the evolution of software methods, since it provides the first “industry standard” object oriented modeling notation, supported by a fairly comprehensive meta-model. It has enjoyed rapid take up by practitioners and tool vendors alike. Initial enthusiasm has been tempered by the experience of implementation and application. In Q3 of 1998, a Task Group was formed under IFIP 8.1 to promote critical review and research into the improvement and extension of UML and related processes. HICSS 34 provides an excellent forum for international discussion related to this topic. Consequently, the mini track was motivated and approved. The topic elicited much interest. Abstracts and drafts underwent a rigorous review process. Ten high quality papers were selected, representing the work of 24 authors from eight countries on three continents. The program begins with two technically oriented papers. “On Transformations from UML Models to Object Relational Databases” by Wai Yin Mok and David P Paper discusses the transformation of UML models to usable and reliable object-relational schemas with static structures as well as dynamic behavior derived from state charts. “Transforming the OOram Three-Model Architecture into a UML-Based Process” by Jes?s Garc?a Molina, Mar?a Jos? Ort?n Ib??ez, Bego?a Moros Valle, Joaqu?n Nicol?s Ros and Ambrosio Toval ?lvarez proposes a method for marrying the three layer architecture with UML inspired by OORAM and IDEA methods. “Problem and Design Spaces during Object-Oriented Design: An Exploratory Study” by Sandeep Purao, Ashley Bush and Matti Rossi exposes subtle human processing characteristics and differences of behavior depending upon task focus while “UML: An Evaluation of the Visual Syntax of the Language” by Stephen Morris and George Spanoudakis is critical of the UML visual syntax and highlights several flaws, but also suggests remedies.B Henderson-Sellers, G Collins and I Graham in “UML Compatible Processes”, contrast the Rational Unified Process and OPEN, both of which make use of the UML notation. RUP's lack of support for the Project Management Body of Knowledge (PMBOK) and iterative incremental development come in for criticism. The paper highlights the many elements beyond notation, which a competent process must address. Next, attention is focused on the UML Meta Model, its shortcomings and suggestions for improvement in: “Metamodelling Approaches for the UML” by Ansgar Schleigher and Bernhard Westfechtel and “Some problems with the UML v1.3 metamodel“ by Brian Henderson-Sellers. Strong suggestions are made for how the meta model should develop going forward into subsequent versions of UML. Of particular interest is the proposal of the first paper for a robust meta model extension capability to support domain specific extensions in a standard way.Terry Halpin, of ORM fame, then suggests “Augmenting UML with fact Orientation”, which can enhance the expressiveness of models and allow the concise and unambiguous representation of business rules and domain knowledge. Verbalization and validation with user communities are promoted. “How to Make Apples from Oranges in UML” by Petri Selonen, Kai Koskimies and Markku Sakkinen discusses transformation (semi-automated) between various UML models, with a view to supporting reverse and forward engineering. Finally, Luciano Baresi, Franca Garzotta and Paulo Paulini address a very topical issue in “Extending UML for Modeling Web Applications” where they borrow from Hypermedia Design Model (HDM) to address the design of dynamic hyperlinked web sites and provide an example.Taken together, the papers present a fascinating insight into how UML is being used, where it could be improved or extended, what some of the alternatives are and how development can be taken forward.The mini-track will encourage a high degree of interaction from attendees and a slot has been reserved at the end for expected lively discussion flowing from the content of the program. We trust that the body of research represented here will flow into the larger UML revision debate and fuel the work of the OMG task force as well as giving interested researchers and practitioners much food for thought. Many thanks to all authors, reviewers and my co-chairs for their very worthwhile efforts.
Graham McLeod, Terry Halpin, Hannu Kangassalo, Keng Siau, "Unified Modeling Language (UML): A Critical Evaluation and Suggested Future", HICSS, 2001, Proceedings of Hawaii International Conference on System Sciences. HICSS-34, Proceedings of Hawaii International Conference on System Sciences. HICSS-34 2001, pp. 3045, doi:10.1109/HICSS.2001.926340