What is Open Source GIScience?
2/28/21
In short, Open GIScience is the practice of keeping source code (in this case, for geographical information science) free and accessible to the public— to both use and contribute to.
Being in this class (Open Source GIScience, GEOG0323) gives us (students and professor) an opportunity to practice open source science in community with one another. Because open source science is so dependent on community for exchanges of ideas, it would be limiting to the field to follow the traditional honor code rules of using only your brain and your professor’s lectures to produce ideas. This dependence on others as an integral part of the field requires radical transparency and credits reported often. While transparency and credits should already be common practice in any field, these practices seem more central and inherent to the ideals of open source work, as it relies on the continuation of these practices over time to ensure reproducibility, allowing for knowledge to grow and ideas to be exchanged.
The increased access and thus interdisciplinarity allowed for by keeping source code free seems particularly relevant to us Middlebury students, as a key principle of the liberal arts education is creating connections between various fields of thought.
While free software and open source are similar movements, “open source” is more about practicality and “free” is more about morality. Moreover, free software is always open source, but open source software is not always free. There are additional complications when open source work is incorporated into commercial software; for example, Google Earth used open source code in its development but now refuses to release its source code (Rey 2009, 204).
The difference between free beer and free speech in the context of open source code is particularly compelling, where “free beer” refers to something you receive without cost, and “free speech” refers to the ability to contribute to the field. This balance between giving and taking provides a framework in which the field can grow by meeting the needs of developers and users, meanwhile also providing recognition for those who contribute the most to encourage progress and adding a sense of fairness. There is also an understanding that those who benefit from open source code will donate to support the cause.
While there seems to be a sense of trust in the open source community, there also exist many different licenses to protect developers’ intellectual property. These licenses can vary but generally seem to add both a sense of protection and hostility, given legal action could come into play. While the apparent necessity of these licenses seems to be at odds with the freedoms defined by the Free Software Foundation (FSF), the FSF itself supports particular licenses created by the GNU. Moreover, the FSF has ironically copyrighted much of the GNU operating system to avoid its privatization.
On a societal level, the benefits of open source GIScience are immense as they reflect an idealized power structure where knowledge, accessibility, and decisions are democratized. This restructuring is in many ways a political statement as it demonstrates a non-capitalist way to operate.
The benefits of open source software also extend practically, as the software exists in a space where it can extend beyond its perceived limits as fields and knowledge grow. Stepping away from non-transparent or unresponsive customer service, users have a community to depend on for fixing issues, or could even fix issues themselves as all the source code is available for their use. This transparency and additional documentation also ensures that users use tools correctly, whereas graphic user interfaces (GUIs) may more easily hide relevant information. And, of course, open source software is free, evening the playing field with powerful corporations and saving costs for users and organizations.
While I think the benefits of open source software outweigh the risks for society at large, I will concede that when thinking about the government and private businesses, the drawbacks of relying on the “little people” may be greater. It would likely require hiring someone who is familiar with open source work and fixing problems themselves by sifting through information, as opposed to calling customer service. While open source software would allow the government or a company to have access to cutting edge software, it also takes extra time and intentionality to download new versions which may have bugs— although long-term releases (LTRs) help mitigate some of this chaos. The “learn as you go” model—keeping up with documentation, new versions, security concerns, etc.— is likely more suited toward scientists than government organizations or private businesses.
Sources not linked above:
Rey, S.J. 2009. Show me the code: spatial analysis and open source. J Geogr Syst 11:191-207. Singleton, A.D., S. Spielman, & C. Brunsdon. 2016. Establishing a framework for Open Geographic Information science. International Journal of Geographical Information Science 30(8):1507 -1521.