DLI Social Science Team

DLI Social Science Team Home Page

send comments or questions to: l-neuma1@uiuc.edu

Object Worlds and Shifting Infrastructure: Building a Digital Library for Engineers

Laura Neumann S. Leigh Star, prepared for March 1996 site visit

We conducted a series of interviews with project members in the late summer of 1995. We reasoned that if we understood the jobs and perspectives of other groups, it would help us understand the designer-user relationship and improve our evaluation efforts. We wanted to understand what it is to successfully construct a piece of information infrastructure. From the worm's eye view, it can be hard to see cooperation and coordination across a project distributed over disciplines and buildings. The interviews gave us a bird's eye glimpse of the emergent Illinois Digital Library, and how this project--which is about federating information spaces--is itself federating project members goals, documents, and timelines in a cooperative venture.

The product of our digital library project, a piece of information infrastructure, differs a bit from more traditional engineering projects in which there is a tangible "thing" result. Infrastructure is not something easily seen as tangible or something that you can put your hands on. Real work goes into negotiating standards, talking with publishers, and translating goals across teams, but this is invisible in the final product. Good infrastructure, in fact, is transparent and invisible (until it breaks down)!

Different people also come to the project with divergent goals, or what we call "side bets." They have different interpretations of what the infrastructure IS, depending on which part of the elephant they are currently touching. And they are at different career stages and have experiences with different aspects of the project, e.g. SGML tagging or database design.

These differences are normal parts of any large project, but because of the unique "invisible" aspect of infrastructure, can seem confusing or disjointed at times. However, ambiguity is necessary to accommodate the multiplicities of needs, world views, and interpretations of the variety of participants in this project. Flexibility and alternative understandings must be facilitated in order for our project to be successful. In practical terms, this means that day to day work practices of individuals should not be rigorously monitored, that different people should be encouraged to hold and share alternative views, and finally, a wide variety of people should be involved in our project. Implications of this are that communication and project unity will be a challenge, but rather than forcing standardization, openness must be maintained. Regular project wide discussions and meetings will be useful to mediate this situation.

In addition to these specific recommendations, this paper contributes to the theoretical understanding of infrastructure and how building it looks in practice. We provide a model of the project, drawing on three analytic concepts in organizational studies:

The design of objects must be both concrete and open-ended (Bucciarelli, MIT);

People juggle multiple agendas and commitments in project work ("side bets") (Becker, Washington);

Both the project components and the people are evolving along complex time-lines ("trajectories") that affect (1) and (2) (Strauss, UCSF).

Finally, we provide some metaphors and analogies to facilitate cooperative discussion of this enterprise of which we are all a part.

NOTE: This paper was presented at the yearly conference of the Society for the Social Studies of Science in Charlottesville, VA in October of 1995.