In this paper, we discuss how multiple input modalities may be combined to produce more natural user interfaces. To illustrate this technique, we present a prototype map-based application for a travel planning domain. The application is distinguished by a synergistic combination of handwriting, gesture and speech modalities; access to existing data sources including the World Wide Web; and a mobile handheld interface. To implement the described application, a hierarchical distributed network of heterogeneous software agents was augmented by appropriate functionality for developing synergistic multimodal applications.

/pdf/multimodal-maps-an-agent-based-approach-1e51o1j1gt.pdf

Multimodal Maps: An Agent-Based Approach

Last year's Rede Lecture at Cambridge by C. P. Snow is now available in a slim, thoughtful, and disturbing volume entitled The Two Cultures and The Scientific Revolution . It runs a mere 58 pages of direct, insistent, pellucid prose and can be read in one hour. Since it takes so little time to courageously tackle two problems vital to scientists, and since it is modestly priced, this unimposing book may well be the greatest literary bargain of an inflated publishing age. The "two cultures" in the title are the modern culture of the scientific intellectual and the traditional culture of the literary intellectual. With a foot in each cultural camp, C. P. Snow comes uniquely armed to the fray. He states his own qualifications: "by training I was a scientist; by vocation I was a writer." More specifically, Charles Percy Snow, the scientist, won his M. A. in physics in

The Two Cultures and The Scientific Revolution

Today's telecommunication systems are enhanced by a large and steadily growing number of supplementary services, each of which consists of a set of service features. A situation where a combination of these services behaves differently than expected from the single services' behaviors, is called service interaction. This interaction problem is considered as a major obstacle to the introduction of new services into telecommunications networks. We present a survey of the work carried out in this field during the last decade (1988-98). After a brief review of classification criteria that exist for feature interactions so far, we use a perspective called the emergence level view. This perspective pays respect to the fact that the sources for interactions can be of many different kinds, e.g., requirement conflicts or resource contentions. It is used to rationalize the impossibility of coping with the problem with one single approach. We also present a framework of four different criteria in order to classify the approaches dealing with the problem. The general kind of approach taken, a refinement of the well known detection, resolution, and prevention categories, serves as the main classification criterion. It is complemented by the method used, the stage during the feature lifecycle where an approach applies, and the system (network) context. The major results of the different approaches are then presented briefly using this classification framework. We finally draw some conclusions on the applicability of this framework and on possible directions of further research in this field.

The feature and service interaction problem in telecommunications systems: a survey

In this special issue of the International Journal of Robotics Research we present a selection of papers that were presented initially at the Sixth International Conference on Field and Service Robotics (FSR'07) held on July 9-12, 2007 in Chamonix, France.

The International Journal of Robotics Research (IJRR) - Special issue on ``Field and Service Robotics ''

Preface This thesis could not have been written without the help and inspiration of many people around me. At the Vrije Universiteit, I was in an interesting environment with many different ideas. I want to thank Gerrit van der Veer, Anton Eliëns and Hans van Vliet for their enthusiasm and how they enriched my research by showing me their perspectives. Thanks to Gerrit for introducing me to the " world of Human Computer Interaction ". I very much enjoyed our collaboration where we had different approaches to a common interest. Thanks to Anton for " convincing " me to do a Ph.D. in the first place and for showing me the use of several interesting techniques during the development of my tool EUTERPE. I also want to thank many other people who supported me throughout the last years. A big part of that was due to Bastiaan Schönhage, by being a friend and roommate for many years. We sure had a lot of fun in between the " serious " work. It was very good to have someone like him to discuss " raw " ideas and to keep me from " overdoing " it. During the last months, we were both writing our dissertation which definitely helped me to keep up the spirit and finish it in time. In addition, thanks to my direct colleagues Nico Lassing, Jacco van Ossenbruggen, Frank Niessink and Jaap Gordijn. During a working day, it is also important to relax and have pointless discussions, for example during lunch when Arno Bakker and Gerco Ballintijn used to join us. Thanks to Hallvard Traetteberg for his work on the patterns. During the development of the patterns he was my best reviewer and contributor. Without his help I would not have developed them so fast. Joerka Deen also proved to be a good discussion partner, always emphasizing the more " philosophical " side of patterns. And of course, many thanks to my parents, my sister and Merche for all their support and love during these years.

/pdf/task-based-user-interface-design-152qropmhj.pdf

Task-based User Interface Design

Tutorial lecture notes on the Irish School of the VDM

The members of Working Group Omega originally belonged to Working Group Beta and, therefore, their deliberations ought to be seen in the context of that working group. Already at Seillac II,2 the issue of the user’s conceptual model (UCM) received serious attention. We felt that the time was ripe to make an attempt at the formal specification of such user conceptual models. As a starting point we take the following definition of a UCM:

Seeheim Workshop on User Interface Management Systems Report of the Working Group on the User's Conceptual Model*

To publish is to make public. And one sense of being public is surely to be accessible? Today it is not only the writing and the images that are published formally, that is to say through official channels, but also the casual human artefacts, the chat, the blog, the quick pic, the self-made music and dance and film, and all of the latter through the medium of the social network. In the World-Wide Web (WWW), to be published is to have a unique resource identifier (URI) and usually a unique resource locator (URL). But to be visibly published on the WWW one needs to be found (much in the same way that one might be found say, 200 years ago, through the library catalogue). Hence at the very core of electronic publishing is to be found the metadata nucleus. In olden times the scholar/reader would have to travel to that place, the Library, if it were accessible, to read/study the work. Today, (s)he travels electronically to those places which are accessible. E-publication does not necessarily entail accessibility. For example, many scholarly works are behind pay walls, costs are borne by institutions of would-be accessors; someone has to pay for maintenance, security, and accessibility. Works of art are in a peculiar and particular category. A work of art is considered to be unique, by which one understands that there is no other copy, properly understood. There may be thousands of prints of the unique piece authorised. But the digitization of an artwork forces a categorical change. The digital artwork is, by nature different. It can be seen, not by reflected light but by transmitted see-through light! In this specific regard it is completely other vis-a-vis the book qua text. In this paper we consider the typical state of the “digital art” as e-publication and explore the extent to which such art is freely accessible to the public, whether on social network or otherwise, with respect to four chosen “National Art Galleries” on the circumference of the European Union.

Social Networks and the national art gallery (Dublin|...|Sofia)

There is still clearly widespread resistance to the deployment of formal methods in industry. On the other hand, certain fields of application, viz. VLSI design, protocol specification, and certain kinds of system, viz. safety-critical, real-time, have of necessity already deployed some form of well-defined engineering methods to ensure a high degree of robustness and reliability in delivered products. In the telecommunications field there is currently a sense of the inevitable in the absolute necessity for the ultimate deployment of formal methods for service specification.

Formal Methods and Service Specification

The act of distributing and the resulting distribution are notions which lie at the kernel of any distributed system. The basic algebra of such distributions and their use in formal specifications has already been developed in terms of indexed monoids (i.e., function spaces with valuations in monoids) and their morphisms. Complementary to such algebra is a body of emerging geometry/topology of formal specifications, one critical aspect of which is the fibre bundle, and more generally the sheaf.

Fibre bundles are used to model the nature and shape of geometrical objects and to associate a field with points in a space. They find particular application in theoretical physics, for example. We demonstrate here that fibre bundles occur naturally in specifications and models associated with formal methods.

/pdf/the-geometry-of-distributions-in-formal-methods-2etqrkobg5.pdf

Mícheál Mac an Airchinnigh

Papers

Tutorial lecture notes on the Irish School of the VDM

Seeheim Workshop on User Interface Management Systems Report of the Working Group on the User's Conceptual Model*

Social Networks and the national art gallery (Dublin|...|Sofia)

Formal Methods and Service Specification

The geometry of distributions in formal methods