To explore the merits of continuous coordination, this project will research, develop, and evaluate a collaborative software infrastructure called Neon, that transcends the traditional distinction between formal and informal coordination. Many problems require collaborative efforts by groups of people and it has been a long-term goal of many researchers to provide software systems to support collaboration. In general, these systems take either a formal or informal approach. Formal coordination is achieved by partitioning work into separate, insulated activities that are periodically shared and resynchronized. Informal coordination is supported by a variety of human activities; often they are ad hoc, but from a general perspective, they are intended to raise awareness of concurrent actions. Neither formal nor informal approaches alone are satisfactory. In particular, the strict isolation maintained in formal coordination frequently leads to integration problems at coordination points and informal coordination has difficulty scaling and often leads to problems of visibility and awareness. These problems compound when confronted with the reality of coordination needs in distributed, collaborative settings: e.g., how to account for the fact that different groups and tasks require different levels of coordination that must continuously be adjusted to support the coordination needs of the situation at hand?
This research parts with two fundamental assumptions underlying approaches to date. First, it breaks the long-standing belief that formal and informal coordination must be treated as opposites, or even that one form is subsidiary to another. Second, it parts with the traditional approach of treating coordination as a strategy that is selected at initiation of an activity and then maintained throughout the duration of that activity. Instead, Neon will follow an integrated and dynamic approach to supporting collaborative work that combines formal and informal coordination and supports fluidly changing levels of coordination. The result, continuous coordination, is highly flexible and continuously adapts coordination support to the needs of the task at hand.
Underlying Neon is a coordination protocol that supports applications with three critical design elements: multiple perspectives, translucency, and non-obtrusive integration. Neon raises both technical and social research questions: (1) how to share, condense, and visualize high volumes of information regarding parallel and potentially dynamic activities, (2) what protocols will support workspace-to-workspace communication and collaboration, (3) what levels of information sharing and collaboration will and will not be tolerated by developers, and (4) how, if at all, will users change their work habits?
The research will empirically evaluate Neon in three application domains: configuration management, software design, and face-to-face, ad-hoc collaborative meetings, which represent a broad range of collaborative settings and requirements. Studies will involve laboratory experiments, controlled experiments in classes, and field studies at NASA/Ames and IBM Research.
Continuous coordination is a novel approach to effective collaboration in complex, dynamic, distributed settings. The principle of continuous coordination is general and not limited to the three examples of configuration management, software design, and face-to-face meetings. Results are likely to transfer to many domains of broad societal impact, such as collaborative learning, collaborative research and analysis, and emergency response.
