Executable and Translatable UML -- How Difficult Can it Be?

Abstract

Model-Driven Engineering is a paradigm for software engineering where software models are the primary artefacts throughout the software life-cycle. The aim is to define suitable representations and processes that enable precise and efficient specification, development and analysis of software.Our contributions to Model-Driven Engineering are structured according to Boyer's four functions of academic activity -the scholarships of teaching, discovery, application and integration. The scholarships share a systematic approach towards seeking new insights and promoting progressive change. Even if the scholarships have their differences they are compatible so that theory, practice and teaching can strengthen each other.Scholarship of Teaching While teaching Model-Driven Engineering to undergraduate students we introduced two changes to our course. The first change was to introduce a new modelling tool that enabled the execution of software models while the second change was to adapt pair lecturing to encourage the students to actively participate in developing models during lectures.Scholarship of Discovery By using an existing technology for transforming models into source code we translated class diagrams and high-level action languages into natural language texts. The benefit of our approach is that the translations are applicable to a family of models while the texts are reusable across different low-level representations of the same model. Scholarship of ApplicationRaising the level of abstraction through models might seem a technical issue but our collaboration with industry details how the success of adopting Model-Driven Engineering depends on organisational and social factors as well as technical.Scholarship of Integration Building on our insights from the scholarships above and a study at three large companies we show how Model-Driven Engineering empowers new user groups to become software developers but also how engineers can feel isolated due to poor tool support. Our contributions also detail how modelling enables a more agile development process as well as how the validation of models can be facilitated through text generation.The four scholarships allow for different possibilities for insights and explore ModelDriven Engineering from diverse perspectives. As a consequence, we investigate the social, organisational and technological factors of Model-Driven Engineering but also examine the possibilities and challenges of Model-Driven Engineering across disciplines and scholarships.

Development of a modeling education program for novices using model-driven development

“…Executable UML is an extension of UML that allows models to be executed and translated into code via model compilers [1]. The basic elements of executable UML are class diagrams, state machine diagrams, and action languages.…”

“…Executable UML is also used to model the component and class diagrams, and the state machines during the design phase, instead of UML. It has been demonstrated that executable UML can be used by novice software modelers [1]. It was also used to introduce the abstract thinking processes involved in modeling, before introducing the more concrete thought processes involved in programming with the standard textual imperative programming languages used at high school level [9].…”

Industrial experiences from multi-paradigmatic modelling of signal processing

“…Ericsson has previously had success in using xtUML for reuse of platform-independent models [1] and test generation [6] while still finding the tool easy enough to use by novice modellers [4]. We also know from experience that xt-UML integrates nicely with legacy code written in C. Since the models are executable it is possible for users to validate that the models have the required functionality without generating code for deployment.…”

“…From our experience of xtUML [4] it does not take that long time to get used to the syntax of the modelling elements, as it does to get used to the tool that creates them. More importantly, as the project evolved we saw a need for handling the data flow in a way that neither xtUML nor the existing platform supported.…”

Mastering model-driven engineering

“…We therefore employed an executable modeling language that supplied the students with continuous validation [3]. Using the terminology of cognitive apprenticeship teacher-student time was used for modelling and coaching while scaffolding and exploration distinguished the work the students did on their own.…”