(Project members only.)
The aim of the MMiSS project, which is supported by bmbf in its programme "Neue Medien in der Bildung", is to set up a multimedia Internet-based adaptive educational system, covering the whole subject of Safe Systems. Thanks to the uniform integration of hypermedia course materials and formal programming tools, teaching in this area will attain a level hitherto impossible in this form. The system will be as suitable for learning on campus and for distance-learning with their associated routine of assignments, as it is for supervised-, co-operative- and self-study.
The system is to be introduced step by step, during the duration of the project, into the normal teaching of the project partners: the University of Bremen, the Distance-University of Hagen, the University of Freiburg, LMU Munich and the University of the Saarland. However, as the "Open-Source" model is to be used and teaching materials and tools are to be made freely available, a much greater national and international take-up is to be expected. To assist this, within the project, a forum is to be founded with German, international, and industrial members, which will advise on what new subjects should be added to the system, and its development and distribution. The forum's advice will be in view of both academic and future industrial applications.
The area of "Safe Systems" has in the last few years become increasingly important. Software is increasingly used to control security-critical embedded systems, in aeroplanes, spaceships, and cars; and electronic trading over the Internet, with its associated security risks, is rapidly expanding. All this requires qualitatively and quantitatively better training in Safe Systems. To go with the planned deployment at universities, a number of well-known German companies have already expressed, through the various industrial contacts of the project partners, an interest in measures for further training in their own workforce.
The core of the system is the translation into hypermedia of a series of classes or lectures on the development of safe systems. The lecturers should be able to store in the system various sorts of course material, such as transparencies, commentary, bibliographies, books, lecture notes, exercises, animations and so, and retrieve them again to use in teaching. The system determines a formal framework for the storage based on an analysis of the semantic structure (ontology) and enables the fast directed access to single teaching elements. An initial collection of teaching materials is already available and should as part of the project be further hypermedially developed. This covers the use of formal methods in the development of (provably) correct software. Highlights include data modelling using algebraic specifications; modelling of distributed reactive systems; handling of real-time with discrete events; and the development of safety-critical systems. The initial collection also covers informal aspects of modelling, and outlines the management of complex projects and security.
The system will also contain a meta-database, containing methodological, ontological and paedagogical information about the contents. The system should, where possible, make the teaching materials available in several different forms. It should be left to the teachers, or even the students, to choose which to use, according to the educational or industrial context. For example reactive systems could be modelled with either process algebras or Petri-nets.
It is important that training should teach the possibilities and limits of formal tools. Tools for formal software development should be integrated in the educational system, to illustrate and deepen the stuff to be taught. Thus students doing assignments can use the system to test their own solutions, while gathering experience of such systems. The integration of a didactic approach with formal methods raises teaching to a new level. It will additionally be possible both to identify common themes in many different tools, and to use them as a new medium for teaching. Thus can for example an algorithm be simultaneously taught, visualised, and verified.
The goal of applying the new system in as many universities and companies as possible, and the fact that secure systems will continue to develop in future, requires the highest level of flexibility, extensility and reusability of the content. It should be possible to incrementally extend or adapt content and metadata, to suit the teacher's own requirements, and to keep them up-to-date.
As the individual parts of the curriculum rely on each other, there is a network of semantic dependencies, which the system should be able to administer; thus it must at the least handle version- and configuration-management. The ontology additionally allows better support for orientation and navigation within the content. It should also form the basis of user changes to the system, for example by learning from exercises which concepts the students have understood, and adapting future assignments accordingly.
The formalisation of semantic dependencies means that the system can help maintain the consistency of the content. Definitions of terms must be adapted to suit each other; the removal or adaption of part of the material may force the removal of adaption of all dependent concepts. Formal software development has similar problems; here also there are semantic dependencies between different parts of development, for example between specification and implementation. Some of the project partners have already developed schemes for the administration of such dependencies as things change, and implemented them in tools for development. Here we perceive an important synergy between expertise in formal software development - and tools for supporting it - and the demands of long-term administration of consistent multimedia materials in competitive educational surroundings.