What is Open Source GIScience?

These days, to say that change happens slowly would merit a concussion test. Over the past year, the entire world has changed to adapt to remote working and learning environments, and the behavior of almost eight billion people has had to shift to adjust to a new normal. Many of these changes will remain a part of modern societies for decades to come, and for better or worse, I as a student will have to live my life and work towards the completion of my goals within the framework of some of these changes. However, there are a few very important fields that I will have the unique opportunity to participate in and not only help advance, but help shape and mold for the future. The most important of these fields, to me, is Open Source GIScience.

While Open Source GIScience isn’t a direct consequence of the 2020 coronavirus pandemic, its unique features are being highlighted and celebrated in the new circumstances of scientific research, geographic information systems, and computer sciences. The basic principles stipulate that computer code used to build software, and even more broadly, scientific research as a whole, must be publicly available in order for any software or scientific experiment to be properly peer reviewed (and therefore, valid). As valid scientific research is the only research that matters, it subsequently becomes very, very important that there is an open source aspect to modern science.

Within the umbrella term “open source GIScience,” there are a few subgroups. Firstly, there is a difference between “free source” and “open source.” “Free source” generally refers to the freest software / code / data one can imagine: not only is it free of charge for one to access and use this code, but it allows the further step of modification, contribution, and distribution. Free source data can be taken by any individual user in the world and made their own. That individual can then, if they wish, spread their creation around their community even further like a bazaar market, sometimes under the protection of a copyleft (a nice little play on words that efficiently reflects the fundamental definition of the word—instead of being bound by law to an existence as property, a product is bound by law to a fully public and accessible existence, for itself and its derivatives).

“Open source,” by contrast, differs in that it is still a product being given out. Open source products are great for practical use, but lack the ability to be contributed to. This inhibits any community from really forming around a product, and therefore bars any of those myriad benefits a community would add. “Open science” stems from this term, but is more friendly because really, free and open are synonyms in the context of “doing science.” Open science seeks to solve the puzzle of valid research through advocating and advancing transparent methodologies featuring accessible and reproducible data banks, workflows, and softwares.

Due to the interdisciplinary nature of geography, GIScience (and everything under that umbrella term) proudly wears the mantle of open science pioneer. GIScience combines computer science, scientific research, crucially relevant conclusions, and even blossoming academic foundations to stand tall as the perfect discipline to accelerate open science. The cultural and legal terms and conditions are being written and rewritten every day with every passing research project undertaken and every new student graduated, so there is a lot of room for individual influence and significant flux in the status quo.

The foundational pillar of everything, as I see, should be the unfailing and dogged attention to attributing credit to every detail of a GIS researcher’s work. If everything is to be public and freely available to read and change, then accomplishments can only be recognized by the collective community. In open science communities, recognition comes in the form of use of a product, and if the whole point of open science is to produce information and tools at a pace and quality like never before, then recognition is crucial. If the scientists necessary to advance the quality and quantity of open-source scientific products can’t make a living do so, they will have to transition into roles creating proprietary tools and software, instead of contributing to open-source communities.

Undergraduate geography departments have a huge role to play in the development of this field. By teaching open source practices, such as thorough attention to detail in workflow and methodologies, the pillars of attributing credit will be second nature to GIScientists when they enter the workforce. Schools are also hubs for experimentation and innovation, and often have the technological and mental resources to test new waters and pioneer yet more scientific breakthroughs. In Open Source GIS here at Middlebury this spring, I am looking forward to sharpening my spatial analysis skills while simultaneously helping build an entirely new grading system for my work to be evaluated in. The opportunities are both educational and exciting.

Finally, there are undoubtedly some significant challenges to the broad adaptation of open science. The largest obstacle is that the world is currently built around supplies and demand, even of knowledge and information. This means that much science is framed in a way that is impossible to truly reproduce—the code isn’t free source, or the data is proprietary, or the publisher doesn’t support a journal article containing a detailed workflow, or the library the researcher is using doesn’t have the infrastructure to house the necessary code. These are all real and significant obstacles to true reproduction and validation of a conclusion, but on the bright side, none of them are physically inalterable. As open science has begun to change the way GIScience is conducted and undergraduate geography is taught, it will steadily change the way data is handled and research is published.

In conclusion, open science is not only new and exciting, but revolutionary and necessary. To work, researchers must accept the responsibility and master the art of attributing credit everywhere and anywhere, for every conceivable detail of a project. Once this can be done, data, research, and knowledge as a whole will shift from being the property of elite intellectuals to a collective feature of an informed, synergistic, and advanced society.

Readings:

Rey, S. J. 2009. Show me the code: spatial analysis and open source. Journal of Geographical Systems 11 (2):191–207. http://dx.doi.org/10.1007/s10109-009-0086-8

Singleton, A. D., S. Spielman, and C. Brunsdon. 2016. Establishing a framework for Open Geographic Information science. International Journal of Geographical Information Science 30 (8):1507–1521. http://dx.doi.org/10.1080/13658816.2015.1137579

To learn more about open source GIS, visit OSGeo.

Submitted March 01, 2021.

Main Page