Research Agenda on Open SourceBrian Kahin, University of Maryland
I submitted a longer background statement for the January workshop. See http://cip.umd.edu/kahin.htm
1. We need research on how the characteristics of open source are actually experienced by users in government and users of government funded software. These include cost (present and future), transparency, security, reliability, adaptability, extensibility, externalities…. How does evaluation relate to circumstance? How does public sector experience and evaluation compare with the private sector experience? How does experience measure up to expectations? What happens when open source replaces proprietary alternatives?
2. What is the actual project-level experience with software patents? What happens when a possible conflict is identified? Do project leaders call attention to patent problems and seek to resolve them? Or do they ignore them? Is there any sense of best practice or means for communicating or coordinating across projects?
How frequent are patent problems? How are they resolved in practice? How do particular resolutions, or lack thereof, affect the progress of the project, including distribution? What warnings or disclaimers are used?
Some companies have argued that inadvertent copyright infringement is a greater risk than patent risk? What is the actual experience with copyright infringement, how is it discovered and dealt with, and how does compare in frequency and serious with patent conflicts.
Do these experiences differ between the U.S. and Europe? (e.g., does greater propensity to litigate and more and broader patents lead to greater problems in the U.S.)
3. How does open source software contribute to productivity in the software sector? There has been long discussion and debate about how software productivity can be measured (e.g., http://www.nber.org/~confer/2002/criws02/wasshausen.pdf), but so far there has been no attempt to take the economics of open source into account. Open source might increase productivity by virtue of the fact that large amounts of functionality can be built on directly, used to develop complementary software, or simply used to displace more costly proprietary software in businesses. (However, this might not be the case if open source software is more expensive to support and maintain.) While this is being addressed as a management at the firm level, it also needs to be investigated at the national level.