Reproducibility and Open GIScience
Open-source GIS not only solves many causes of the reproducibility crisis but also has other added advantages for individual researchers who adopt the ethos and the broader geographic science community as a whole.
The obvious way that open-source GIS enables reproducibility is that making methods and data freely available invites other researchers to reproduce a study. Encouraging reproducing and replicating studies could lead to a wider culture shift in academia of checking peers’ work to increase our breadth of knowledge without the fear of being disproved. The implication that this will have on journal and publishing culture is similar to the implication I believe incorporating open-source GIS into education will have on grades. Open-source data and methods (i.e. code) eliminates the need to “get published”. Without this competition to publish work before other colleagues, there is no reason to be proprietary over one’s work. Similarly, an open-source learning environment in college eliminates an environment that fosters competition with peers for grades and performance. I think our work in this course can still be evaluated and labeled with a grade despite the open-source nature, despite what the Honor Code might assert. This grade may in fact be a better reflection of our ability to problem solve and our understanding of GIS than a grade based on correctness of a final product. Researchers of Open GIScience benefit from a similar shift–instead of being “graded” by which journal you are published in and how many citations you get, their projects benefit from the collaboration of many people to be as effective or thorough as possible.
One advantage of open-source GIS that is salient to me as an individual is the ability to always have access to your software, as opposed to being limited on a licensed computer or associated with an institution. It’s a little counterintuitive to think that you can only “own” your GIS work when it’s open-source.
I’ve been thinking a lot about the role open-source GIS could have in government because this is the other environment that I have used GIS in outside of academia. In my experience (specifically within land-management agencies), government agencies tend to shy away from Open Source software of all kinds due to potential security risks. However, government agencies are often cost limited and simultaneously can’t afford sufficient GIS licenses or the most recent GIS software. This results in not being able to fully harness the tool of GIS or even just use the bare minimum functions of GIS. I think government entities have embraced GIS by making their collected data more available to the public and for free, but they still have yet to embrace the open-source ethos to improve internal work. I was surprised to learn about the Biden-Administration’s commitment to open-source science and reproducibility because the government agencies I have personally witnessed were stuck at least five years in the past when it came to technology. I think government adoption of open-source tools should be a priority, especially for land management agencies who have a great need for GIS but often little funding. Furthermore, these are agencies that work with less sensitive information than, for example, the defense sector (who happens to have much more funding and can afford the latest and greatest in GIS).
References
NASEM. 2019. Reproducibility and Replicability in Science. Chapter 3: Understanding reproducibility and replicability, pg. 31-43. Washington, D.C.: National Academies Press. DOI: 10.17226/25303
Rey, S. J. 2009. Show me the code: Spatial analysis and open source. Journal of Geographical Systems 11 (2):191–207. DOI: 10.1007/s10109-009-0086-8