1 Towards Automated Testing of Web Usability Guidelines Dominique Scapin and Corinne Leulier Dominique.Scapin@inria.fr, Corinne.Leulier@inria.fr Institut National de Recherche en Informatique et en Automatique Unité de recherche Rocquencourt, Domaine de Voluceau – B.P. 105, F-78153 Le Chesnay Cedex, France Web: http://www.inria.fr/Dominique.Scapin Jean Vanderdonckt and Céline Mariage vanderdonckt@qant.ucl.ac.be, mariage@qant.ucl.ac.be Université catholique de Louvain (UCL), Place des Doyens, 1 – B-1348 Louvain-la-Neuve, Belgium Web: http://www.qant.ucl.ac.be/membres/jv Christian Bastien Christian.Bastien@ergo-info.univ-paris5.fr Laboratoire d'Ergonomie Informatique, Université René Descartes, 45, rue des Saints-Pères – F-75270 Paris Cedex 06, France Web: http://www.univ-paris5.fr/LEI Christelle Farenc, Philippe Palanque, and Rémi Bastide Farenc@univ-tlse1.fr, Philippe.Palanque@univ-tlse1.fr, Remi.Bastide@univ-tlse1.fr Université Toulouse I, UFR d'Informatique, Laboratoire d'Interaction Homme-Système (L.I.H.S), Place Anatole France – F-31042 Toulouse, France Web: http://lis.univ-tlse1.fr/~farenc/ , http://lis.univ-tlse1.fr/palanque/, http://lis.univ-tlse1.fr/bastide/ 1. I ntroduction Usability of web sites, whatever their type is (e.g., internet, intranet, extranet, supranet) is today widely recognized as a major dimension required for user acceptance and use of web sites. This dimension even becomes more critical for electronic commerce web sites: a customer unsatisfied by poor usabil- ity is likely to become a competitor’s customer. Moreover, the user population is expanding in age (ranging from young users to elderly people), in expectations (ranging from private use for leisure to professional use), in information needs (ranging from simple information to compound multimedia re- sources), in task types (ranging from basic text searches to complex problem-solving methods), and in user abilities (ranging from the able-bodied person to any person with special needs, such as for mo- tor-, auditory- or visually-impaired persons). Therefore, web sites should be usable enough to accom- modate all these variations over time. Numerous evaluation methods help ensure that web sites are usable (Nielsen,94; Nielsen & Mack,94): inquiry methods, inspection methods, user testing, to name a few categories of them. These methods are so numerous and hard to differentiate that identifying which ones are suitable for a particular case study is challenging. This identification is often practically constrained by several needs, as:  The need for a cost-effective method: it is often heard that project resources in time, budget, and personal are so limited that they prohibit any use of evaluation method (Lynch, Palmiter & Tilt,98).