In our view, a collaborative system (which we will call a Collaboratory) provides an open electronic platform for individuals or groups with common interests to efficiently exchange, disseminate and create issues, ideas and knowledge. We use the field of MIS as an illustration of our approach to conceptualizing and designing a Collaboratory. However, the ideas presented are general enough to be applied to virtually any discipline. Unlike a proprietary system or application, which provides only a few customization choices to users through limited application programming interfaces, an open system such as the Internet provides a wide range of design choices, which require a careful consideration on part of the developers.
We believe that if properly organized conceptually and technically, the Web can serve as a highly effective foundation for a Collaboratory, which will bring together the diverse worlds of researchers and professionals in various fields. On the other hand, isolated and idiosyncratic developments of Web servers will lead to information overload, inefficiency and chaos. We argue that ``smart agents'', which are sometimes regarded as a panacea for the Infoglut and the information filtering problems, will be less effective when information is organized in a haphazard manner. In addition, without a pragmatic yet sound theoretical basis for assessing the value of interactions among various researchers and professionals, we cannot get a sense of which direction to pursue in terms of future technological developments on the Web. In this paper, we use complementarity theory [2, 3], an old but powerful concept from economics, to develop a model linking a Collaboratory user's net value to the system design features. It provides a rational foundation for understanding the drivers of the value derived by the users of a Collaboratory, and for selecting design characteristics in such a way that maximizes net user value. In the context of Collaboratory design, an application of complementarity theory would suggest that when a Collaboratory can provide a set of related (complementary) features through a single, open user interface, it greatly enhances the benefit to the user while reducing his/her cost of effort and time spent in using the system, thereby maximizing the user's net value. Complementarity theory would also imply that having only a subset of complementary design features in isolation will not bring the desired benefits to the Collaboratory users.
With this perspective, we develop a prototype MIS Collaboratory, which is a Web server with information organized according to our vision of the MIS field. The Collaboratory also incorporates an on-line electronic forum, announcement and search features. The MIS Collaboratory has two main objectives: (i) efficient dissemination of information through organization of information resources, search and resource linking capabilities, and (ii) a global forum for asynchronous and synchronous interactions involving MIS issues, ideas and research articles. We report our experience with the usage of the Collaboratory, as well as some summary statistics of data collected from world-wide participants on the electronic MIS Forum.
The contribution of this paper is twofold. First, it provides a theory-based foundation for designing open, global Collaboratories with the objective of maximizing user value. Given the large number of design alternatives in an open environment, it helps rationalize why certain features of a collaborative system must be provided in tandem to increase user benefits and reduce opportunity cost. Second, it uses the theoretical premise to design and implement what we believe to be the first Web-based interactive electronic forum.
In section 2, we review the relevant prior literature and provide the motivation to develop a foundation for designing computer supported collaborative systems. A complementarity theory based framework for a Collaboratory is presented in section 3. Section 4 focuses on the design of a Collaboratory, while section 5 provides implementation details. Some preliminary experience with the Collaboratory and summary statistics of data collected from Collaboratory users are presented in section 6. We discuss one possible architecture of a global collaborative system involving multiple Collaboratories in section 7. Future research directions and concluding remarks are provided in section 8.