The Science of Collaboratories (SOC) project at Michigan is an effort to mine the large experience base that has emerged with building collaboratories to support geographically distributed scientific work. Our first two workshops explored the social and technical issues at a very general level. The next stage was to get more detailed, and to compare and contrast various collaboratories to begin to build some principles "bottom up" about what makes collaboratories successful. We are doing this at two levels of analysis: a broad survey of some aspects of all the collaboratories we can identify, and a more in-depth analysis of a smaller set. Our third workshop in September 2002 was our first foray into this strategy.
First, we have an effort we call Collaboratories-at-a-Glance, where we are creating an inventory of every collaboratory project we can identify that has been carried out. Of course, this hinges on what we mean by a collaboratory. We have been using the following definition:
A collaboratory is a network-based facility and organizational entity that spans distance, supports rich and recurring human interaction oriented to a common research area, fosters contact between researchers who are both known and unknown to each other, and provides access to data sources, artifacts and tools required to accomplish research tasks
As we gather examples of collaboratories in this "at a glance" effort, we are relaxing some of the conditions in this definition so that we create a broad set of examples that will help us refine our definition as well as understand the issues pertaining to collaboratory success.
The collaboratories-at-a-glance effort gathers a small set of basic facts about each collaboratory we find. But we want to investigate some collaboratories in much greater detail, so the second task we're pursuing is this. In order to refine this deeper investigation, in September we selected four collaboratory projects for initial investigation. These four were:
- Upper Atmospheric Research Collaboratory (UARC)/Space Physics and Aeronomy Research Collaboratory (SPARC).
- Great Lakes Regional Center for Aids Research (GLR CFAR).
- Environmental Molecular Science Lab (EMSL) collaboratory.
We gathered materials about these four projects guided by an extensive framework we created based on a number of hypotheses about the factors that contribute to collaboratory success. We read papers, studied web sites, and interviewed project principals for each of these projects.
Since September we have identified six other projects that we are investigating in depth:
- Alliance for Cell Signaling (AfCS).
- Biomedical Informatics Research Network (BIRN).
- Grid Physics Network (GriPhyN).
- International Virtual Data Grid Laboratory (iVDGL)
- Network for Earthquake Engineering Simulation (NEESgrid)
At the June workshop we are updating the Collaboratories-at-a-Glance effort, and then presenting details about these six projects. We then revisit the important theme of cross-cutting issues that we began at the September workshop.