Socially-Oriented Requirements Engineering: Software Engineering Meets Ethnography
We outline an approach for eliciting, understanding, and representing the cultural aspects of the domestic environment for the purpose of system design. We use agent models as shared artefacts to represent the everyday cultural life of the home. These rep
- PDF / 493,399 Bytes
- 20 Pages / 439.36 x 666.15 pts Page_size
- 12 Downloads / 179 Views
Socially-Oriented Requirements Engineering: Software Engineering Meets Ethnography Sonja Pedell, Tim Miller, Frank Vetere, Leon Sterling, and Steve Howard
11.1 Introduction Technology can facilitate interpersonal contact in social interactions, but only if it addresses and fulfils the felt needs of people acting in their social contexts. The felt needs include those that are emotional or behavioural, such as experiencing playfulness, feeling engaged, or being capable of expressing intimacy (Howard et al. 2006; Leonardi et al. 2009; Vetere et al. 2009a; Vutborg et al. 2010; Yarosh et al. 2009). Such socially-oriented requirements are important to human culture but are difficult to specify and measure. Consequently, engineering systems to fulfil them is a nontrivial task. The functionality needed to facilitate a socially-oriented requirement is often unclear; for example, how do we engineer a system to ensure it is fun? Ethnographic data can be used to inform system models and to help define socially-oriented requirements (Martin and Sommerville 2004; Viller and Sommerville 2000). However, ethnographic data does not translate into requirements in a straightforward manner. Themes extracted from ethnographic data are not functional requirements (Randall et al. 2007). Ethnographies are rich descriptions of human activities and cultural practices, and do not define the
S. Pedell () Faculty of Design, Swinburne University of Technology, 144, High Street, 3181 Prahran, VIC, Australia e-mail: [email protected] T. Miller • F. Vetere • S. Howard Department of Computing and Information Systems, The University of Melbourne, 3010 Parkville, VIC, Australia e-mail: [email protected]; [email protected]; [email protected] L. Sterling Faculty of Information and Communication Technologies, Swinburne University of Technology, 3122 Hawthorn, VIC, Australia e-mail: [email protected] V. Dignum and F. Dignum (eds.), Perspectives on Culture and Agent-based Simulations, Studies in the Philosophy of Sociality 3, DOI 10.1007/978-3-319-01952-9__11, © Springer International Publishing Switzerland 2014
191
192
S. Pedell et al.
behaviour of technological systems. Ethnographic data tends to be a bottom-up view of the domain, while system models are typically derived top-down. Albeit critical, informing system models with ethnographic data remains challenging. This chapter defines a method for addressing the gap between ethnographic data and system models created with agent-oriented techniques. We argue that the agent paradigm (Sterling and Taveter 2009) is suitable for modelling the social domain because it allows representation of the goals and motivations of agent roles and individuals. By social domain, we mean those practices that encompass cultural activities and embrace shared values. Specifically, we argue quality goals can be used to discuss socially-oriented requirements such as having fun and being playful. Our method substantiates and refines agent-oriented quality goals with attributes and new understandings abou
Data Loading...
