Accessibility stands as a quality requirement for Web applications. However, current accessibility automatic evaluation tools are not capable of evaluating DOM dynamic generated content that characterizes Ajax applications and RIAs -Rich Internet Applications. In this context, this paper describes an approach for testing accessibility requirements in RIA, by using acceptance tests. The authors had implemented a set of assistive technology user scenarios in the acceptance tests, in order to guarantee keyboard accessibility in web applications. As the scenarios were implemented as acceptance tests scenarios, they provide accessibility analysis over all layers of the software, from server-side to client-side implementations (JavaScript and dynamically generated DOM elements) in RIA. The test scenarios are automatically executed, and by doing so, fit the Continuous Integration process of constant delivery of new functionalities in Web projects.

No abstract

The Web is currently the main way of providing computing services, reaching a larger number of users with different characteristics. As the complexity and interactivity of systems is increased, users become more demanding towards all the requirements associated to their distinct needs. Implementing the interaction requirements in the Web has become the main focus of accessibility and usability studies, describing essential design features which provide users with quality, assured systems. The focus on the users reinforced that as the number of users grows and the system became available to a wide variety of users, accessibility and usability features become even more critical to a Web application's success. In this paper, we present ACCESSA, a practical approach to rapidly improve the accessibility of existing Web systems, acting mainly in the interface design with no changes to the functional requirements of systems. The ACCESSA is based on the WCAG 2.0 guidelines and other patterns, choosing the guidelines that present lower implementation costs and represent higher severity accessibility issues.