Modern science is increasingly collaborative. But while collaboration has always been a part of science, those who collaborated in the past were often collocated. Today, science needs to be able to take advantage of specialized talent available regardless of location.
However, anyone who has participated in a project where participants are scattered across different locations knows that distance matters (Olson & Olson, 2000). Given that, how can we increase the chance that scientific collaborations across distance will succeed, and what does it mean for a project to be "successful"? These questions motivated a 5-year, NSF-funded investigation known as the Science of Collaboratories (SOC) project. The goal of the SOC project was to define, abstract, and codify the underlying technical and social mechanisms that lead to successful distributed collaborations.
One of the major outcomes of the SOC project is the Theory of Remote Scientific Collaboration (TORSC). TORSC resulted from our attempt to distill basic theoretical issues from the host of best practices and lessons we learned over the course of the SOC project and from literature on computer-mediated communication, organizational behavior, management information systems, and science and technology studies. TORSC proposes a broad set of success measures and analyzes factors that affect those measures. (A forthcoming book from MIT Press entitled Scientific Collaboration on the Internet includes a chapter on TORSC as well as case studies and research findings from investigations of distributed scientific collaborations, including a chapter on LTER coauthored by Bill Michener and Bob Waide.)
Factors that lead to success
Five major clusters of components are important to success: the nature of the work, the amount of common ground among participants, participants' readiness to collaborate, participants' management style and leadership, and technology readiness. The major categories, with the exception of the management issues, were first described in Olson and Olson (2000). We have since identified the key management and decision-making practices that are critical to success and detailed the significant components within these clusters. Table 1 lists the key factors and examples of each. Many of these factors affect collocated collaborations, but are critical to distributed projects.
Table 1: Factors that lead to success
Nature of the Work
- Participants can work somewhat independently from one another.
- The work is unambiguous.
Common Ground
- Previous collaboration(s) among participants was successful.
- Participants share a common vocabulary, and if not there is a "dictionary" to help with translations.
- Participants share a common management or working style.
Collaboration Readiness
- The culture is naturally collaborative.
- The goals are aligned in each sub-community.
- Participants have a motivation to work together. For example, the participants like working together; there is something in it for everyone; and/or the project requires a mix of knowledge and skills.
- Participants trust each other to be reliable, produce high quality work, and have their best interests at heart.
Management, Planning, & Decision Making
- The principals have time to do the work.
- There is a critical mass of people at each location.
- There is a point person at each location.
- Management, communication, knowledge organization, and data sharing plans are in place. Plans have room for reflection and redirection.
- Leadership sets culture and develops management plan.
- No legal or financial issues remain such as intellectual property rights.
Technology Readiness
- Collaboration technologies provide the right functionality, give benefit to the participants, and are reliable and easy to use.
- Participants are comfortable with the collaboration technologies.
- Technical support is available at each location and an overall technical coordinator is in place.
Forms of success
There are many ways for a distributed collaboration to succeed, and different sets of factors may lead to different kinds of success. TORSC identifies five categories of success: the science itself; science careers; learning and science education; funding and public perception; and the development of new collaborations. For example, a remote collaboration may succeed by encouraging the use of new tools to communicate and coordinate across distance, diversifying the pool of people who become scientists, improving the satisfaction of those in the field, and ultimately, providing revolutionary breakthroughs in both the conduct and outcomes of science.
Conclusion
In TORSC, we have identified the major factors that appear to be important for successful distributed research collaborations. Further research is needed to illuminate the logical connections between the factors, and to identify which factors are the most significant and under what circumstances they apply.
At its core, TORSC states that revolutionary science results when scientists work collectively and diverse points of view are brought to bear on a common problem. Technology then kicks in by allowing more diverse and distant groups of scientists to communicate with each other so that their collective work is coordinated. But major tensions also arise from having large and diverse groups of scientists working together, and the tendency in such groups to have less common ground, lower degrees of trust, and the need for stricter coordination and management. By facing these tensions and focusing on the factors that affect positive outcomes, we expect to see an increase in successful remote collaborations in the future.
Reference
Olson, G. M., and Olson, J. S.. 2000. Distance matters. Human Computer Interaction 15:139-179.