No abstract

Avatars and anthropomorphic characters by marketers are becoming more commonplace on commercial web sites. Moreover, a trend among marketers is to use ethnically ambiguous models in advertising to appeal to specific consumer segments. This study helps our understanding of not only how best to segment and appeal to racially diverse consumers but how people interact with virtual human agents in relationship to the literature on audience response to real humans. It was predicted that Blacks would respond more positively to a Black agent, than they would to either a White agent or an ethnically ambiguous agent. It was also expected that Whites would show no difference in their response based on the race of the computer agent. The findings demonstrate that Blacks had more positive attitudes toward a computer agent, had more positive attitudes toward a web site and recalled more product information from a site when the site featured a Black agent vis-à-vis a White agent. Whites showed no significant response difference concerning the agent, the brand or the site based on the racial composition of the computer agents. Interestingly, the ethnically ambiguous character was overall just as effective in persuading both White and Black browsers as were the same-race agents.

Ubiquity in Information Systems (ISs) is a new requirement widely expressed by customers and users due to emerging and evolving communication and mobile technologies. Each IS should support a set of mobile applications used either to interact smartly with the changing environment, to provide adaptive services to customers or both. Designing ISs with highly technological risks requires a precise and appropriate development process. However, such processes fail to consider ubiquitous requirements throughout the development process. This paper tries to solve this issue by proposing a process for identifying and modeling ubiquitous requirements that can be integrated into an existing IS engineering process. This process, called E-CARe, focuses on adapting to the surrounding context; this requires detailed specification and analysis work by a context designer. E-CARe uses an event-driven logic, as dynamicity and reactivity are the major properties required from ubiquitous applications. A Model-Driven Engineering (MDE) approach is used to automate specification work. In order to test the process, a case study from the intelligent transport domain is applied as an illustration.

We present ParTraP and its associated toolset, supporting a lightweight approach to formal methods. In critical systems, such as medical systems, it is often easy to enhance the code with tracing information. ParTraP is an expressive language that allows to express properties over traces of parametric events. It is designed to ease the understanding and writing of properties by software engineers without background in formal methods. In this tool demonstration, we will present the language and its toolset: compiler, syntax directed editor, and a prototype generator of examples and counterexamples .