Francis Jambon

Using the B Formal Approach for Incremental Specification Design of Interactiv Systems

This paper introduces a new technique for the verification of interactive systems. It first presents the use of a model oriented formal method for specifying interactive systems, i.e. the B method. Then, it suggests formally based solutions which allow solving difficulties that are inherent to interactive systems specification, like reachability, observability or reliability. Our claim is that this model-oriented technique that uses proof obligations can be used together with model checking techniques, where automatic proofs of properties can be performed.

Dialogue Validation from Task Analysis

Up today, formal methods have mainly been used to allow designers to verify that software conforms to its specification. In this article, we propose a validation method and a tool to analyse whether the design actually fulfils the original requirements for the system. The principle of our validation method is to generate the complete set of possible user interaction sequences from the task analysis. Then, this set is injected in the Dialogue Controller Component of the application. At last, the Dialogue Controller Component’s calls to the Functional Core are intercepted, and compared with the user’s goals. Our case study is in the general Computer-Aided Design area, in which systems support a huge number of tasks.

Formal modelling of task interruptions

My doctoral research is concerned with the formal modelling of task interruptions. Although interruptions are significant events in human activities, current models and notations do not support their expression appropriately. My contribution to this problem is twofold: the ISAU model which makes explicit the general structure of an intermptiow and a UAN-based formal notation that would force designers to consider the right questions when developing a system. ISAU will be assessedusing a real-world exemplar: the Data-Link system that supports communications between pilots fmm diiTerent airmaft’s and air tmftlc eontmllem

Formal Verification and Validation of Interactive Systems Specifications

This paper proposes a development process for interactive systems based both on verification and validation methods. Our approach is formal and use at first the B Method. We show in this paper how formal B specifications can be derived from informal requirements in the informal notation UAN. Then, these B specifications are validated using the data oriented specification language EXPRESS. Several scenarios can be tested against these EXPRESS

Meta-Evaluation of a Context-Aware Mobile Device Usability

This article analyzes the completeness and the correctness of usability analyses conducted in usability laboratories compared to analyses conducted in reality testing (where the context is real and video recordings impossible for technical or privacy reasons). To do so, we set up a meta-evaluation of a mobile device in a quasi-realistic context. We collected two sets of data. The first type refers to data usually available in a classic usability laboratory, the second simulates data which could be collected in reality testing. We did two independent analyses and then, afterward, compared their findings. The meta- evaluation concludes that it is possible -with limitations- to evaluate the usability of a mobile device in reality testing without video recordings. However, the usability problems detected may remain unexplained without the help of the video recordings.

Conception participative par « moments »

This paper describes an approach for the participatory design. After describing the trends in the domain, a formalization of the design process is expressed based on «phase», «moment» and «primitive». The phases articulate the design moments as nodes and transitions in a dynamic graph. The moments are built on reusable primitives defined as instanciable objects. Then, the paper details with this formal model a platform and tools for managing the design process on different points of view: from the user and engineers to the project manager.

First steps in the retro-engineering of a GUI toolkit in the B language

This paper provides a progress report on an initial approach of retro-engineering of a graphical toolbox using the B method. Nowadays, dependable systems which feature graphical user interfaces do not use Widgets libraries, safety of which was ensured. The use of formal methods (the B method in our study) is an interesting solution making it possible to guarantee some of the properties (safety and conformity) of these Widgets. We show that retro-engineering of classical Widgets (Button, Slider) or more elaborate ones (RangeSlider) is possible but error prone and reveals some badly controlled concepts.

Gaze-based interaction: evaluation of progressive feedback

In monomodal approaches, eye-tracking for gaze-based interaction suffers from a tight coupling between perception and action: making the distinction between user action and user perception of information is almost impossible. This paper proposes the concept of progressive feedback to release this coupling. First experiments confirm that gaze-based interaction can be credible in some contexts of use. Moreover, progressive feedback appears as possibly valuable.

Plateforme d'utilisabilité/usage MultiCom: objectifs, réalisation, usages et retours d'expérience

This article describes the design approach and experience gained when implementing an original usability laboratory: the MultiCom platform. The purpose of this platform is the evaluation of the usability and usage of a wide range of interactive systems: from classic systems to ubiquitous environments. For this, the platform was designed as a large theater where acts could be played. The relevance of the design choices (surface, ground plane, automation) was validated through several experiments that were successfully carried out on the platform.

L’identification radiofréquence au service de l’analyse du parcours muséal des visiteurs

Analyser le parcours des visiteurs tout au long d’une exposition n’est pas chose aisee. Les auteurs montrent comment la technique de l’identification radiofrequence (RFID) peut apporter une aide significative a cette analyse et quels en sont les resultats, les limites et les perspectives.