According to recent studies an estimated 90% of Web sites and applications suffer from usability and/or accessibility problems. As user satisfaction has increased in importance, the need for usable Web applications has become more critical. To achieve Web usability for a Web product (e.g., a service, a model, an application, a portal), the Web artifacts' attributes must be clearly defined. Otherwise, assessment of usability is left to the intuition or the responsibility of people who are in charge of the process. In this sense, usability models (describing all the usability sub-characteristics, their attributes and the relationship among them) should be built and Usability Evaluation Methods (UEMs) should be used during requirements, design and implementation stages based on these usability models. Another major challenge in industry today is to make the Web more usable for everybody, including those with disabilities. Since there is no general method for evaluating the usability and accessibility of a Web application in all circumstances, the workshop can contribute to clarify several concerns such as the following:

1. What are the shortcomings of existing approaches for Web usability and accessibility evaluation?
2. How the Web application usability can be decomposed into a set of measurable attributes? For example, the ISO 9126 standard decomposes usability into five sub-characteristics. The standard does not specify how these sub-characteristics can be effectively and efficiently measured.
3. What is the suitability of existing UEMs to evaluate Web interfaces?
4. What are the specific usability criteria for Web artifacts? To which extent do they depend on the family of applications under consideration?
5. How can we broaden accessibility evaluation to involve people with disabilities? Does it provide a more effective solution than merely conforming to accessibility standards?
6. What are the strengths and weaknesses of tools for automated evaluation? For example, many accessibility tools are based on accessibility guidelines, but vary in their interpretation of guidelines. So these results still require human judgment.