"The Thing is... What is Our 'What'?"

This paper presents an ethnographic EM/CA study of a seven-hour meeting of an information system design team working on the design of an information system intended for use by multiple systems. Facilitating system interoperability is an important challenge currently facing design teams. Our study elucidates—in the team's own words—the problems they confront with "object" and "concept" certainly, practical "use" issues and "language". Human use and comprehension frame their concerns. If the team defines data objects loosely in an attempt to make concepts universally shareable—human users will create multiple private extensions to achieve more certainty with the effect that the one system they design will functionally become many. If they define data objects tightly to avoid this, their data objects may lack the ability to be shared, requiring expensive recoding or translation of data as it moves across boundaries. The team refers to these as "technical", "political" and "philosophical" issues with defining a "What". There are also social dimensions to the practice-related issues that arise when "Things" defined as conceptual objects in one data schema must be used across information systems by human workers. We highlight the design issues raised by the team and elucidate several social aspects of those issues.