Event Summary | Low-Cost and Open Source Tools: Next Steps for Science and Policy

State of the Field: Open Hardware and the Open Science Agenda

“We live in a time that we can really see, particularly from the current pandemic, how important it is to have timely and accessible access to scientific data publication and information,” said Ana Persic, Chief of Science Policy at the United Nations Educational, Scientific and Cultural Organization (UNESCO), who set the tone during her keynote address. 

But despite the recognized potential of open science, no international standard for open science exists. UNESCO is leading a multi-stakeholder process to draft an international Recommendation, which Dr. Persic describes as an “international framework for open science policy and practices that is cognizant of the regional differences in open science perspectives.” Key to the development of the UNESCO Recommendations is “understanding the opportunities and challenges” related to the complex, yet equally essential parts that compose open science, including open hardware. 

Following Dr. Persic’s keynote, a robust panel discussion explored open hardware’s value across scientific disciplines, and for both traditional and open science paradigms. “If we're not open sourcing the hardware designs behind the results in the published literature, we can't reproduce [the results],” said Dr. Zac Manchester, Assistant Professor of Robotics at Carnegie Mellon University and founder of the KickSat project, who also discussed how open hardware reflects and supports science’s core value of reproducibility. With open hardware, scientists can better analyze how data is produced, and even replicate the tools themselves when validating a research result.

“I would definitely second that with regards to reproducibility-- one of the great things about open hardware,” agreed Dr. Meghan McCarthy, Program Lead for 3D Printing and Biovisualization at the Office of Cyber Infrastructure and Computational Biology at the National Institute of Allergy and Infectious Diseases (NIAD). She also noted “my PhD is in neuroscience, but...had I had access to open hardware, the experiments that I could have done, or ways that I could've done it...more efficiently, is really exciting to me.” According to Dr. McCarthy, open hardware can “make experiments faster and better and advance time to discovery” due to the ability to build on the work of others.

Dr. Gerald “Stinger” Guala, Program Scientist at National Aeronautics and Space Administration (NASA)’s Earth Science Division, discussed the collaborative environment low-cost and open source tools can generate with open licenses and other enabling structures for grassroots collaboration. He illustrated this point by sharing an example of how two competing citizen science projects (and NASA grantees) “organically” decided to work together in this enabling environment: “I would love to take credit for the collaboration, but I didn’t do it.”

Because open hardware can be shared among scientists and iterated upon, Dr. Guala argued for a high return on investment, particularly for government funders. “We can fund the piece of hardware once, and then everyone else just uses that and does something better the next time.” 

Infrastructure emerged as a critical topic during the panel discussion. Both Dr. McCarthy and Dr. Guala touched upon how repositories are infrastructure that can help avoid duplication of designs (a challenge also mentioned by Dr. Persic) while increasing  scientific relevance. For example, the NIH 3D Print Exchange--founded in 2014, and debuted at the Obama Administration’s White House Maker Faire--is a research repository that allows any interested party to submit 3D printing blueprints for review, and gained popularity during COVID-19. Despite the platform’s success, Dr. McCarthy also spoke of challenges, including a lack of standards to exchange blueprints between different repositories to “maximize the value of government funding” and help communities share their work. 

Dr. Manchester spoke of how access to standards enabled the small satellite community: “if it meets the spec, you can go launch it.” Across science and technology domains, standards can provide a benchmark for evaluating attributes like quality and establishing fitness-for-purpose. According to Dr. Manchester, they also enable collaboration by providing a focal point for researchers to converge around. 

Approaching the theme of infrastructure from a complementary angle, Dr. Guala considered the link between big-data producing hardware and software infrastructures. One of Dr. Guala’s projects is a forthcoming cloud-based data analysis platform built in cooperation with the European Space Agency (ESA) and others. Dr. Guala noted that NASA--like other granting agencies--traditionally funds researcher access to and participation in open science infrastructures. However, while requirements for the cloud-based analysis platform “are built around open science principles,” NASA is still trying to figure out how to offer broader public access in a cost-efficient way. 

Simply opening or providing access to a resource isn’t sufficient to ensure that it can be used. “What's really important is not just getting the data out there, but making ways that it is accessible” explained Dr. McCarthy, “I'm really interested in helping to make this kind of data adhere to FAIR principles of being Findable, Accessible, Interoperable, Reusable,” a framework for thinking through these complexities. 

Interoperability is a technical question, and a legal one. As Dr. Guala points out, licencing is a prerequisite to cooperation, because it enables the legal and ethical sharing of devices. And, as Dr. McCarthy explained, the challenge isn’t just using an open license, but also “understanding the differences between copyright and patent, which is really big in hardware.”

Documenting open resources using metadata, including but not limited to information on licensing, is an important starting point. But as Dr. Manchester pointed out, metadata alone “doesn’t get at why it is the way it is or how it really works.” He tried to address this challenge by documenting the rationale behind the design process and sharing test results. However, numerous panelists agreed that there are no established guidelines for how to accomplish this in practice, let alone a protocol with strong community adoption. 

Possible Futures: Elevating Impact on Science and Society 

Shannon Dosemagen, a Shuttleworth Foundation Fellow with the Open Environmental Data Project, offered closing remarks. Drawing on over a decade experience in open hardware communities, Ms. Dosemagen made a compelling argument for “why open hardware needs policy attention now.” This argument can be further explored in the Journal of Open Hardware’s Medium Series examining opportunities and recommendations for open hardware in science and policy. 

Her rich recommendations included brokering connections between open source networks and government chains, investing in building block technologies, and supporting data sharing frameworks. However, when asked to distill these into “one thing” an incoming Administration should focus on, Ms. Dosemagen didn’t hesitate: “our trust in science is broken and our trust in many of our governing systems is broken. We have a real opportunity to try to figure out ways to forge new bonds between science and society. Thinking about how we broaden inclusivity, broaden participation, that's really at the top of my mind right now.”

Low-cost and open hardware can help bridge this gap by aligning grassroots and governments perspectives while collaborating on the key challenges facing science and society. At the Wilson Center, our hope is to keep offering a platform for communities to come together and take the next steps forward. If you’re interested in supporting these efforts, don’t hesitate to get in touch: Dr. Anne Bowser, anne.bowser@wilsoncenter.org.

Resources