The following is a placeholder for pictures of the presenters in action
The following is a placeholder for pictures of the BBQ @ Eugene's home
The following is a placeholder for pictures of the Saturday 29 August 2009 MSDL canoe trip to Parc La Mauricie
The diversity of today's model transformation languages makes it hard to compare their expressiveness and provide a framework for interoperability. De-constructing and then re-constructing model transformation languages by means of a unique set of most primitive constructs facilitates both. We thus introduce T-Core, a collection of primitives for model transformation. Combining T-Core with a (programming or modelling) language enables the design of model transformation formalisms. We show how basic and more advanced features from existing model transformation languages can be re-constructed using T-Core primitives. Finally, we design a new primitive language for model transformation by combining T-Core with an existing asynchronous and timed modelling language.
presentation [pdf]11:00 suprise visit by Prof. Manfred Nagl
In this talk we present an overview of the main challenges and achievements regarding systematic testing of model transformations. We focus on rule-based (graph) transformations as the main representative of model transformations. The talk is divided in to two sections with different emphasis. The first half focuses on a high level overview of the main activities required to achieve automated testing for model transformations, using an analogy to the well established field of system testing in software engineering. We will discuss the work done, how it contribute to the overall solution of this complex problem and finally what challenges and activities that we still need to address to provide a complete solution. The second half of this talk will focus on the Model Differencing challenge. A survey of the recent attempts to tackle this problem is categorized and presented. We investigate the possible connection to the sub-graph isomorphism problem and forms of it, in an attempt to adapt and re-use the main achievements and optimizations, to address the different but related problem of model differencing.
presentation [pdf]For more than a decade, the importance of various forms of modeling in the software engineering process has been steadily increasing. Widely accepted standards such as the Unified Modeling Lan- guage (UML) have been at the forefront of most modeling efforts. In recent years however, it has become apparent that such standards merely offer more appealing views of the systems to-be-built but still remain quite close to the final implementations. Domain-Specific Modeling (DSM) aims to model systems using problem-domain concepts much closer to client requirements than conven- tionally used solution-domain concepts thus truly raising the level of abstraction far beyond that of code. There is no shortage of skeptics regarding DSM's capability of replacing current software development practices due in part to the great many open problems that stand in the way of DSM being applicable to real world large scale problems. This presentation will motivate DSM in more detail by providing the historical background that led to its creation and it will cover the important open problems as well as the past and current research they have spawned.
presentation [pdf]Metamodeling provides a unified way to describe various domains. The predominant metamodeling languages have been developed to describe the structural information of models. However, they are not designed to encode the behavioral information. As the executability is essential for simulation and validation, designers have to either refer to extensions or resort to programming languages to describe the actions. This fact results in two shortcomings. On the one hand, lack of executable model prevents including complete model information in the model; and on the other hand lack of neutral language makes transferring models between tools and transforming models between formalisms non-trivial. In this project, we designed an self-contained metamodeling language, ArkM3 which adds the definition of neutral action language to the meta-meta-level. We also rebuilt the kernel of AToMPM, now called "ark", based on this meta metamodel.
presentation [pdf]Statecharts form a fundamental basis for modelling system behaviour. Hierarchical in nature, they allow a modeller to visually represent complex systems, including concurrent behaviour. Following the UML2.0 standardization, with the integration of class diagrams to depict the system structure, a new compiler is provided to translate such a visual representation into some executable computer programming language. In this talk, we will focus primarily on the code synthesis of statecharts, but will also deal with multi-formalism modelling whereby a statechart, usually a continuous-time based model, is translated into a behaviourally equivalent DEVS model, a formalism that can be used as a semantic domain for differential algebraic equations, timed model transformations, etc... The ever-famous DigitalWatch exercise will be used as a simulation example!
presentation [pdf]There are many challenges that developers face during the development of a complex User Interface (UI). Desired behaviour may be autonomous or reactive, and possibly real-time. Each UI component may be required to exhibit a radically different behaviour from that of any other component and the behaviour of components may be inter-related. These complex behavioural relationships between components are often difficult to express, and are even more difficult to encode and maintain. We suggest that a developer should specify the behaviour of a UI using a representation which minimizes "accidental complexity". Unfortunately, conventional code-centric approaches fall short. Our solution uses Model-Driven Engineering (MDE). By modelling every aspect of the system-to-be-built, at the most appropriate level of abstraction, using the most appropriate formalism(s), it becomes possible to completely capture the behaviour of a UI, to rapidly generate prototype implementations, to easily adapt the UI as project requirements change, and, finally, to synthesize a UI and maintain it. More specifically, in this presentation, we introduce a class of UIs known as Scoped User Interfaces, and illustrate how one may model them using Hierarchically-linked Statecharts (HlS). The techniques described in this presentation have been used to develop AToMPM, an experimental, SVG-based visual modelling environment. The presentation will give a step-by-step description of our approach by means of a concrete example, in order to illustrate the work that has been done to develop, using HlS, an SVG-based environment for visually editing and simulating Statechart models.
presentation [pdf]Traditional methods to increase simulator efficiency focus on run-time performance prediction and subsequent allocation of computational resources. Modelers do however hold the most precise information to determine model activity. We propose a Resource-Aware simulation framework based on DEVS. "Hints" about resource useage are represented attached to Domain-Specific models. With the help of model transformations, the hints are transformed to be the performance prediction model in DEVS. Supported the prediction model, a Resource-Aware simulator is built to be capable of Activity Tracking and Activity Prediction. The resource is subsequently balanced during the whole simulation.
presentation [pdf]Advances in communication technology have opened the door for many new types of services, such as the remote delivery of healthcare. This new form of service delivery, not only contributes to the democratization of healthcare, by reaching far-away populations, but also makes it possible for elderly and chronically-ill patients to have their health monitored while in the comfort of their homes. Despite all of these advantages, however, patients are still resisting the idea of medical telemonitoring. One of the main obstacles facing the adoption of medical telemonitoring, is the concern among patients that their privacy may not be properly protected. We address this concern, and propose a privacy-preserving telemonitoring protocol for healthcare. Our protocol allows patients to selectively disclose their identity information, and guarantees that no health data is sent to the monitoring centre without the patients' prior approval. The approval process can be automated, and requires only an initial configuration by the patient.
presentation [pdf]The SpacecraftLib library has been developed in Modelica for use in the domain of Systems Engineering for space systems with a special emphasis on modularity, usability and ease of modification and expansion. It is a multidisciplinary tool which combines all the relevant subsystems. Power, command and data handling, and mechanical models are integrated into a single Modelica device in order to model as completely as possible the behaviour of a physical onboard device. We will describe the tool, examine a case study and briefly analyze the results of a simulation.
presentation [pdf]Quick analysis of thermal systems without the use of CAD models may be very valuable for rapidprototyping. Such a need led to the heavy modification of the Heat Transfer package found in the Modelica standard library. The library in question revolves around a three-dimensional generic block composed of a variable number of interconnected elements with individually assignable thermal and physical properties. When applying the accompanying library functions, one may create moderately complex thermal structures with non-uniform thermal properties. Furthermore, other tools available also allow the user to attach chains of blocks of different resolutions, as well as insert one block into another to create composite models with the possibility of internal heat generation.
presentation [pdf]