To provide an effective solution for CIS, we spent significant efforts on the design phase. This section provides details about how we derived design specifications for Coagmento, built a preliminary interface, tested it with pilot runs, obtained feedback using cognitive walkthroughs, and enhanced the design specifications using participatory design sessions.
Personal Interviews
As Grudin (1994, p.93) pointed out, “many expensive failures in developing and marketing software that is designed to support groups are not due to technical problems; they result from not understanding the unique demands this class of software imposes on developers and users.” Such views are reaffirmed by recent works such as (Shah, 2009). Keeping this in mind, we interviewed a number of people who work and/or teach in information intensive domains, asking questions about their past and present collaborative projects. The details of this study are reported in (Shah, 2010a). Here we will report only those findings that are relevant to the system design.
We discovered that email and face-to-face meetings are some of the most popular methods of collaboration. These methods represent two extremes of the classical model of collaborative methods (Rodden, 1991; Twidale and Nichols, 1996), where email fits on the remote and asynchronous end, and meetings fit on the co-located and synchronous end. However, due to the changing structure of work environments and habits (people working on multiple projects with different sets of collaborators, across multiple sessions, and with multiple devices), the need to fill in the gap between these extremes is more apparent than ever.
While most of the respondents wished for better tools for collaboration, they agreed they would have a hard time departing from familiar tools, such as email and IM, even though these tools were not explicitly designed to support collaboration. Even if people know about tools such as del.icio.us, they still send website links to each other over email. System designers and developers face a grave implication due to this fact; they need to provide seamless integration of tools that support collaboration within a user’s existing working environment rather than making him choose between his tried-and-tested method and a new tool. This finding reflects the views of Grudin (1994), where he suggested extending an existing single-user system, with which a user is already familiar, with groupware features to minimize the cognitive load and maximize the adoption rate. As one of the respondent admitted in our interviews, “We focus on results, and not how to do it.”