Integrative and Comparative Biology Integrative and Comparative Biology, volume 62, number 4, pp. 1061–1075 https://doi.org/10.1093/icb/icac043 Society for Integrative and Comparative Biology SYMPOSIUM Open Hardware in Science: The Benefits of Open Electronics Michael Oellermann *,†,1 , Jolle W. Jolles‡ , Diego Ortiz § , Rui Seabra¶ , Tobias Wenzel || , Hannah Wilson# and Richelle L. Tanner** ∗ Aquatic Systems Biology Unit, TUM School of Life Sciences, Technical University of Munich, Mühlenweg 22, D-85354 Freising, Germany; † Institute for Marine and Antarctic Studies, Fisheries and Aquaculture Centre, University of Tasmania, Private Bag 49, Hobart, TAS 7001, Australia; ‡ Centre for Research on Ecology and Forestry Applications—CREAF, Campus UAB, Edifici C. 08193, Bellaterra Barcelona, Spain; § Instituto Nacional de Tecnología Agropecuaria—INTA, Estación Experimental Manfredi, Ruta 9, Km 636, 5988 Manfredi, Córdoba, Argentina; ¶ CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos, InBIO Laboratório Associado, Campus de Vairão, Universidade do Porto, 4485-661 Vairão, Portugal; BIOPOLIS Program in Genomics, Biodiversity and Land Planning, CIBIO, Campus de Vairão, 4485-661 Vairão, Portugal; || Institute for Biological and Medical Engineering, Schools of Engineering, Medicine and Biological Sciences, Pontificia Universidad Católica de Chile, Vicuña Mackenna 4860, Macul 7820244, Santiago, G92Q+26, Chile; # Biology Department, College of Science, Utah State University, 5305 Old Main Hill, Logan, UT 84321, USA; ∗∗ Environmental Science and Policy Program, Chapman University, 1 University Drive, Orange, CA 92866, USA From the symposium “Open source solutions in experimental design’’ presented at the annual meeting of the Society for Integrative and Comparative Biology virtual annual meeting, January 3–February 28, 2022. 1 E-mail: michael.oellermann@utas.edu.au Synopsis Openly shared low-cost electronic hardware applications, known as open electronics, have sparked a new open- source movement, with much untapped potential to advance scientific research. Initially designed to appeal to electronic hob- byists, open electronics have formed a global “maker” community and are increasingly used in science and industry. In this perspective article, we review the current costs and benefits of open electronics for use in scientific research ranging from the experimental to the theoretical sciences. We discuss how user-made electronic applications can help (I) individual researchers, by increasing the customization, efficiency, and scalability of experiments, while improving data quantity and quality; (II) sci- entific institutions, by improving access to customizable high-end technologies, sustainability, visibility, and interdisciplinary collaboration potential; and (III) the scientific community, by improving transparency and reproducibility, helping decouple research capacity from funding, increasing innovation, and improving collaboration potential among researchers and the pub- lic. We further discuss how current barriers like poor awareness, knowledge access, and time investments can be resolved by increased documentation and collaboration, and provide guidelines for academics to enter this emerging field. We highlight that open electronics are a promising and powerful tool to help scientific research to become more innovative and reproducible and offer a key practical solution to improve democratic access to science. Introduction and Martinez-Fuentes 2018). However, so far one do- The revolutionary open science movement has helped main that is at the very core of scientific data production to foster transparency, collaborative access, and shar- has been less prominent in the open science movement: ing of scientific knowledge (Vicente-Saez and Martinez- hardware, electronics, and instruments (Harnett 2011; Fuentes 2018). Open science started with open-access Pearce 2012; Maia Chagas 2018). Proprietary instru- publications and has now expanded to liberate access ments support high-profile research, such as microflu- to data, programming code, and even lab notebooks idic instruments for transcriptomics or autonomous (Boulton et al. 2012; McCray et al. 2018; Vicente-Saez underwater vehicles to monitor marine environments, Advance Access publication May 20, 2022  C The Author(s) 2022. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 1062 M. Oellermann et al. yet high costs limit their access only to well-funded labs. Most researchers globally do not have access to the funding required to buy state-of-the-art proprietary instruments, limiting both reproducibility and innova- tion potential (van Helden 2012). Free and open-source hardware (Pearce 2013) has the potential to close this divide: in many instances, it can match high-end perfor- mance of proprietary instrumentation while facilitating the sharing of free-of-cost design blueprints to re-build, modify, or advance instruments, and fostering collab- oration with other scientists and a worldwide commu- nity of “makers,” civic scientists, and hobbyist inventors (Pearce 2012; Maia Chagas 2018). Electronics are a major component of the open hard- ware domain, which provides open-design scientific hardware solutions (Pearce 2012; Bonvoisin et al. 2020). Such solutions are often built in combination or solely with particular electronic hardware components whose main purpose is to allow non-experts to easily create or reproduce electronic applications. Therefore, in con- trast to conventional electronics, we define “open elec- tronics” as applications and designs that are assembled from very accessible components and openly shared to facilitate access, learning, and reproducibility at min- imized costs. This contrasts with commercial applica- Figure 1. (A) Cumulative growth of Web of Science records tions, and proprietary or closed designs, which restrict grouped by the top 12 countries and (B) dominant subject areas. access and use of designs by license ownership. To clar- The search comprised the open-electronics boards listed in Fig. 3 and included articles and proceeding papers between 2010 and ify, many open electronics rely on components that 2020 and the country origins for authors and co-authors. For are not open-source and may in fact may be closed- detailed analysis, bibliographic data, country distribution, and design (e.g., the Raspberry Pi [Raspberry Pi Foun- separate analysis for proceedings articles only, see Supplementary dation, Cambridge, United Kingdom]) or commercial Information File S1. (e.g., the DS18B20 sensor [Maxim Integrated, San Jose, CA, USA]), but they are well-suited to create afford- able, accessible, and broad electronics applications. Ap- umentation and components being commonly avail- plications and their designs should be openly published able, and with help of numerous open online tutori- and shared, which, depending on their novelty and als and databases (e.g., instructables.com; hackster.io), complexity, can range from a short description in the open-electronics projects have become very accessible methods or supplementary material section of a pa- to the broader public. With millions of hobbyist “mak- per; online repositories (e.g., open-neuroscience.com, ers” and DIYers around the globe—for example, already Maia Chagas 2021); detailed step-by-step instructions >37 million Raspberry Pi microcomputers have been in methods articles or websites (e.g., Geissmann et al. sold since its release in 2012 (RaspberryPiFoundation 2017); to publications entirely focused on describing 2020)—the popularity of open electronics has contin- tool development and application potential (Plum and ued to rise and is beginning to establish in diverse scien- Labonte 2021). Components dedicated to the open- tific domains, including the biological sciences (Fig. 1A electronics concept include single-board microcomput- and B; Jolles 2021a). ers and micro-controllers that can be easily interfaced Despite their increasing uptake in science, open- with a plethora of hardware modules, sensors, and electronics applications are far from being widespread. actuators (Table 1), and often without much techni- Poor awareness, inadequate documentation, and insuf- cal experience or highly specialized tools. In combina- ficient electronic literacy outside the engineering and tion with the modular nature of many open-electronics computer sciences have contributed to its fragmented platforms, users do not need to invent applications and uneven use across scientific subjects (Fig. 1B). from scratch and can gradually grow skills and applica- In comparison, an open-source software project such tion complexity. By only requiring basic programming as the R statistical language (R Core Team 2021) had and electronics skills in most instances, detailed doc- similar challenges at the start but has now become one Open hardware in science: the benefits of open electronics 1063 Table 1. Overview of the wide range of sensors and actuators available that are compatible with open-electronics boards and platforms Sensors Environment Temperature, humidity, barometric pressure, soil moisture, particulate matter, light intensity, smoke, dust, radiation, sound Water Chlorine, pH, depth and pressure, liquid level, flow, turbidity Gas CO, CO2 , alcohol, H2 , TVOCs, ozone, H2 S, CH4 , NO Movement Distance, acceleration, seismic, GPS, break-beam, motion Biometrics Heart rate, muscle activity, fingerprints, weight/load, force Imaging Spectroscopy, visible and IR range cameras, thermal imaging, gestures Other Magnetism, capacitive touch, current, voltage, RFID, PIT tags Actuators Light LEDs, infrared, UV, laser Movement Servos, stepper motor, gear motor, vacuum pumps, valves Switches Mechanical, electrical, magnetic, DC and AC relays Other Vibration, sound, ultrasound, Peltier heating/cooling of the most popular data tools in science (Muenchen Application potential for open 2012; Lai et al. 2019) due to its high flexibility and rapid electronics adaptation to new research demands and trends via Open electronics offer a versatile spectrum of applica- user-driven innovation networks (Von Hippel 2005, tions to a wide range of users in science, education, and 2007). To reach broad-scale uptake of open electronics, industry, besides the general public. Although initially important barriers need to be overcome by increasing used only by the most electronics-savvy hobbyists and access to detailed information, improving educational Do-it-Yourself creators, open electronics are increas- and research support, and by presenting a clear case ingly taken up by broader public audiences spanning all for how open electronics can benefit researchers, insti- age groups, further fueled by the rise of the Internet of tutions, and the scientific community alike. Increased Things (Ibrahim et al. 2015). Automation of scheduled awareness and use across the biological sciences will tasks such as watering plants in the garden (Divani et advance discussions to more complex barriers, such as al. 2016) or controlling household devices (i.e., smart constrained funding or resources, user-centered design, homes) are very popular and easy to set up (Hasan et or remote location challenges. Open electronics can al. 2018). This extends to various measurements and aid solving some of those barriers with their low cost, surveillance applications, such as automated weather easy to build designs, and better global accessibility. stations and nest box systems, and even the develop- Such actions will help to adopt open electronics more ment of smart cities (Costa and Duran-Faundez 2018). broadly and accelerate technical innovation, lower re- Some of the driving forces behind the rise of open elec- search costs, enable highly customizable solutions, and tronics are the aim to make computing and electronics democratize hardware access in experimental science accessible to anyone, such as in STEM education, and to (Pearce 2015, 2016). introduce students to electronics and programming ba- In this paper, we outline the broad benefits that open sics as well as solving practical problems and practicing electronics can have from the level of the researcher the scientific method (discussed in Jolles 2021a). The to that of institutions and the scientific community at increasing interest in open electronics as teaching tools large. We also discuss current barriers and trade-offs, is supported by an extensive pool of learning resources provide a “beginner’s toolbox” to help researchers get for teaching and self-learning (see Table 2). This also started, and conclude with an outlook discussing poten- supports scientists to try out and test new ideas, learn tial impacts on science and required actions for a broad new approaches, or develop own customized applica- uptake. Overall, it is not our aim to persuade researchers tions on a small budget. to use open electronics in their work but to provide So far, despite much potential, there has only been a comprehensive account and raise awareness of their a marginal uptake of open electronics in biological re- multi-level benefits as a very accessible technological search and science generally, with predominant use in solution for experimental inquiries, as we believe open the engineering and computer sciences (Fig. 1B). How- electronics have great potential to enhance the innova- ever, also in the biological sciences, open-electronics ap- tion, reproducibility, and democratization of science. plications have increased in number over the last few 1064 M. Oellermann et al. Table 2. Collection of online resources and communities, for both beginners and advanced users, with hyperlinks Resource Link Description Arduino website arduino.cc Many tutorials, forums, blog posts and products for sale Biomaker website biomaker.org No-code programming tutorials for biologists Raspberry Pi website projects.raspberrypi.org Many tutorials, forums, blog posts and products for sale Raspberry Pi Beginner’s Guide Free guidebook for getting started with Raspberry Pi magpi.raspberrypi.org/books/beginners-guide-4th-ed Adafruit website learn.adafruit.com Thorough documentation and tutorials for Adafruit products Sparkfun website learn.sparkfun.com Thorough documentation and tutorials for Sparkfun products PiHut website thepihut.com/blogs Thorough documentation and tutorials for PiHut products Raspberry Pi Guide raspberrypi-guide.github.io A collection of 30 + Raspberry Pi tutorials specifically written for scientists (Jolles 2021a) Coursera Coursera.org Offers courses on topics related to electronics and computing Udemy Udemy.com Offers courses on topics related to electronics and computing TinkerCad tinkercad.com Lets you build virtual versions of circuits to test your wiring and code Raspberry Pi Forums raspberrypi.org/forums 300k + member forum to ask questions about Raspberry Pi Open Science Hardware forum.openhardware.science Forum to ask questions related to open hardware and global Forum community events Stack Overflow stackoverflow.com Forum to ask questions about hardware or coding Open-Neuroscience open-neuroscience.com Database for scientific open-hardware designs Github Githubpages Site to create a free, version controlled online website with your documentation Wildlabs.net wildlabs.net Conservation technology network Conservation X Labs conservationxlabs.com Technology and innovation company working against extinction Open Source Ecology opensourceecology.org Ecology-relevant solutions and community around an open-source economy Many companies that sell components for open electronics provide thorough documentation and tutorials. Furthermore, there are guides specifically developed for scientists wanting to work with the Raspberry Pi or Arduino (e.g., Jolles 2021b), and an increasing number of online courses are available on topics related to electronics and computing. The links above are arranged by relevance—starting with beginner tutorials and ending with ways to share applications. years, with numerous solutions found both in tradi- interaction using an Arduino-controlled LED array tional journals as well as newly established open hard- (Johnson et al. 2016) (Arduino, Somerville, MA, SUS), ware journals (e.g., Journal of Open Hardware , Hard- and soil ecologists developed a “smart electronics nose” wareX). to monitor soil organic richness using Arduino-read gas Some examples from the field of behavioral ecology sensors and IoT ZigBee boards to stream data from the include video recording, tracking, and remote solutions, field to the lab (Dorji et al. 2017). such as the high-throughput automated recording of the In evolutionary biology, open electronics were used behavior of individuals and groups of fish (Jolles et al. to perform high-throughput experiments to auto- 2017); in-situ analysis of zooplankton phototactic be- matically simulate and identify adaptive niches of havior underwater (Lertvilai and Durand 2020); ”bi- yeast (Wong et al. 2018). These examples impressively ologging” of migratory and nesting birds (Bridge et al. demonstrate the versatile and fast-growing application 2019; Youngblood 2020); and the analysis of kangaroo potential of open electronics in biological research. Be- rat escape behavior (Schwaner et al. 2021). sides numerous applications in other scientific disci- At the ecosystem-scale, solar-powered Raspberry Pis, plines, they are also formidable tools for community cameras, and microphones have enabled the fully au- science and scientific outreach activities, such as school tonomous acoustic and visual monitoring of ecosystems student-operated ocean observers (Beckler et al. 2018), with remote data transmission for <$320 USD (Sethi urban air pollution monitors (Jiang et al. 2016), in- et al. 2018), and the long-term visual, high-frequency teractive autonomously moving lights to teach about monitoring of forest canopy cover (Wilkinson et al. animal behavior (Jolles 2021, pers. comm.), and sonic 2021). Experimental biologists tested the impact of light kayaks to monitor underwater soundscapes (Griffiths spectral quality and intensity on plant host–parasite et al. 2017). Open hardware in science: the benefits of open electronics 1065 Research benefits of open electronics toxicological assays (Steffens et al. 2017), RFID-based In addition to their diverse application potential, open animal feeding stations (Bridge et al. 2019), or acoustic electronics can provide a broad range of significant sensor networks generating high-density data streams benefits at the different levels of academia and resolve from the field to the cloud (Roe et al. 2018). Task au- important practical, financial, and structural issues in tomation also helps reduce human error and experi- science. Below we discuss the key benefits for individ- mental variability (Eggert et al. 2020) and increases re- ual researchers, for research institutions, and for the silience to unforeseen circumstances, by using auto- academic community at large. mated monitoring systems to alert researchers when equipment fails or experimental conditions change un- expectedly (Gurdita et al. 2016). Benefits to individual researchers Wide applicability, from simple to complex Customization Unlike most commercial scientific instruments, which Most instruments, such as hand-held meters, data log- confine customers and applications to product lines, gers, and PCR machines, are closed entities, constrained open electronics are highly flexible and adaptable and to the functions set by the manufacturer and operat- can be implemented in a broad range of applications, ing software, and can thus become redundant if re- from basic to the highly complex, such as simple video search needs change. The poor ability to modify or ex- recording of snakes (Zamore et al. 2020), advanced IoT- pand functionalities also confines the scope and im- based bird monitoring nest boxes (Abdelouahid et al. plementation of new research ideas. Open electronics 2020), or complex real-time virtual reality systems to provides one solution, as researchers can not only de- study chemotaxis in flies (Tadres and Louis 2020). Users velop or retrofit existing open-electronics setups and can start simple and expand their devices with increas- devices, exchange, or program new operations, but also ing programming and electronics skills, like starting link and expand the features of existing laboratory in- with only logging lab temperature, then displaying it struments. For example, micro-controllers and single- live on an LCD screen, controlling heaters to regulate board computers can interface with proprietary instru- temperature to, finally, a complete stand-alone system ments via serial ports and hardware communication with multiple sensors, warning messages, and interac- protocols, to query information or execute functions, tive graphical user interfaces (GUIs). Users can also eas- while adding new functionalities using sensors and ac- ily repurpose open electronics by reusing components tuators (Stewart et al. 2017; Virag et al. 2021). As many from previous setups for new or more complex builds. boards offer wireless connectivity with an increasing fo- cus on Internet-of-Things applications (e.g., NodeMCU Broad sensor and actuator application potential ESP32 [Espressif Systems, Shanghai, China], Fig. 3), A major strength of open electronics is the wide range even simple weighing scales can integrate into a smart of sensors and actuators available that can be controlled instrument network, channeling and summarizing data with the accuracy of reference equipment (Table 1; streams in cloud-based dashboards (Poongothai et al. Setyowati et al. 2017). Open electronics can also be 2018; Arunachalam and Andreasson 2021). Micro- used in applications requiring an especially small size scopes have especially benefited from this customizabil- (Shipley et al. 2017), both in the lab and under harsh ity, leading to many high-specification or low-cost open conditions in the field (e.g., Beddows and Mallon platforms (Hohlbein et al. 2021). 2018). Micro-controllers and single-board computers also enable multiple sensors and actuators to be con- Scalability and high throughput nected simultaneously, providing a sensing and reactive Open electronics provide researchers with the oppor- capacity that, in many instances, can outperform com- tunity to easily scale and replicate setups to suit sin- plete proprietary solutions while having significantly gular or high-throughput applications. Their low cost lower individual hardware unit needs, costs, and power and off-the-shelf availability enables quick and low- consumption. risk prototyping up until a well-functioning setup that can be copied to create whole arrays of identical de- Experimental automation vices, such as to GPS-track tens of animals (Foley and Repetitive tasks, such as control and recording of ex- Sillero-Zubiri 2020), to test the behavior of hundreds perimental parameters in the field or lab, animal feed- of individual flies (Geissmann et al. 2017), to observe ing, and monitoring of experimental trials, are amongst the growth of thousands of plants (Tausen et al. 2020), the most time-consuming factors in research. Open and to simulate multi-dimensional environmental con- electronics can benefit researchers by automating such ditions by parallelized automated processing of sam- tasks, including by using pipetting robotics in eco- ple microvolumes (Wong et al. 2018). Such scalability is 1066 M. Oellermann et al. particularly valuable when funding is limited, enabling creasing number of open online courses (e.g., Coursera researchers to begin with simpler setups, rather than and Udemy) offer many ways to learn about open elec- facing high upfront costs for proprietary systems. tronics and how to build custom applications. Academic papers now often come with supplementary guides Flexible data access and programming capabilities and accompanying websites about methodologies (e.g., Open electronics are highly flexible in terms of data Geissmann et al. 2017; Maia Chagas 2021), and several acquisition, formats, storage, and accessibility. Numer- specialized journals exist to help researchers build and ous libraries in a broad range of programming lan- publish their own devices (e.g., Journal of Open Hard- guages make it possible to read sensor data in a few lines ware (Murillo and Wenzel 2017) and HardwareX). It of code. Library-rich easy-to-read programming lan- is also easier to troubleshoot problems, as most open- guages such as Python, further facilitate endless possi- electronics applications are built on similar and wide- bilities to work with custom electronics and devices, in- spread building blocks (i.e., Arduino platform) that cluding automatic data processing actions such as folder share a common programming language, and large on- monitoring (e.g., watchdog library), file conversion, and line communities exist that can be consulted to help automatic creation of data backups (e.g., BackuPy li- solve specific problems (e.g., stackoverflow.com and brary). Data can also be accessed remotely, including raspberrypi.org/forums, with >300k members). from a local network and the Internet, and from remote field locations via mobile network adaptors (e.g., RPyC Transferable skills library, Sethi et al. 2018). This, in turn, enables the con- Besides providing practical benefits, learning to work tinuous real-time remote monitoring of data, such as of with open electronics and creating custom devices and lab conditions, animal activity, plant growth, and envi- applications also provides researchers with transferable ronmental variables in the field (Jolles 2020; Siregar et skills, including knowledge of programming and elec- al. 2017; Trasviña-Moreno et al. 2017). Improved com- tronics, technical problem solving, and creative think- puting power of single-board computers has made it in- ing, which is paramount to scientific progress. This new creasingly possible to process data onboard, enabling skill set may spark new and cross-disciplinary ideas the transmission of flagged or summarized data for re- for research. It can further generate engineering- or searchers (Allan et al. 2018). Data can also be visual- programming-related skills, which improve researchers’ ized in a professional manner via custom-build user in- employability outside academia or lead to unexpected terfaces or online dashboards, supported by numerous business opportunities (e.g., the start-up ElectricBlue). graphical libraries, many of which are open-source (e.g., It also enables researchers and institutional staff to be- Tkinter, PyQT, WxPython, Dash, and Plotly) (Boudoire come instrument generalists, who, once they obtain et al. 2020; Lewinski et al. 2020). open-electronics skills, can operate, service, or mod- ify different types of open-electronics setups with no Simple maintenance or less training than for specialized proprietary instru- Most components of open electronics can be easily ser- ments. Technicians and staff can then more flexibly viced and replaced by the users themselves, with most work within and across departments and become more parts likely available at online retailers and electronic valuable for institutions. hardware stores. Also, required tools, such as soldering equipment and a multi-meter, tend to be highly afford- able. In contrast, when issues occur with proprietary Benefits to departments and institutions (scientific) instruments, custom repairs, even when fea- Access to customizable high-end applications sible, are not recommended as they break product war- Access to cutting edge scientific instrumentation is key ranty. Researchers therefore rely on manufacturers for to the success of many research institutions. How- repairs, which can be time-consuming and expensive, ever, expenses to purchase, maintain, or modify high- specifically for regions distant from industrial and trade end instruments are often prohibitive. Open electronics hubs, where return shipments or service staff visits are have advanced to cutting-edge scientific instrumenta- prohibitively costly. Product support may also cease if tion and provide the added benefit of being lower cost products become outdated or companies stop existing. and highly customizable. There are numerous exam- Therefore, open electronics aid researchers to depend ples of open-electronics instruments with uncompro- less on exclusive vendors’ support and be more resilient mising quality being used for high-end scientific re- against potential technical problems. search, such as automated microbiological incubators to study adaptive evolution (Wong et al. 2018), high- Extensive learning resources and community support throughput tracking and optogenetic stimulations to Extensive learning resources, including a large range of study fly’s sensorimotor behavior (Tadres and Louis books and free tutorial websites (see Table 2), and an in- 2020), and microfluidic single cell sequencing prepa- Open hardware in science: the benefits of open electronics 1067 ration (Stephenson et al. 2018). Academic institutions Improved collaboration and visibility can benefit from open-electronics solutions as a leaner Open electronics can facilitate collaborations within way to perform workflows in-house through a mod- and across institutions through the collective develop- ular, gradual investment, overcoming the need for re- ment and implementation of open-electronics solutions searchers to depend on large grants or follow-up fund- for frontier research applications. Institutions can fos- ing for instrumental maintenance and upgrades. Insti- ter this by intra- and inter-institutional think-tanks, tutions can foster uptake by providing dedicated open- workspaces, and shared educational programs, techni- electronics workspaces, where researchers can imple- cal support as well as attributing dedicated time to re- ment ideas, and build institutional networks to share searchers. The diverse application range of open elec- knowledge, ideas, and instruments across departments tronics enhances cross-disciplinary collaboration po- and stimulate interdisciplinary innovation. tential but requires shared benefits for all parties. If successful, collaborations with, for example, engineers may enable highly complex and broadly useful open- Cost reductions and improved sustainability electronics applications. Resulting publications, where When encouraged as an institutional-wide policy, the useful tools are published in addition to research data, cost effectiveness of open electronics can be extended will likely have higher impact and citation rates simi- for the lifetime of the equipment. By supporting re- lar to open data (Colavizza et al. 2020). A clear com- searchers to communicate and document about their mitment to technologies that democratize science will open-electronics instruments and providing common also help institutions to enhance collaborations between guidelines, institutions can enhance the knowledge pool industrialized and emerging nations and attract re- among staff and thereby help generate and support a searchers that can easily cross-transfer open-electronics “user-maker” community. This in turn not only facili- technologies. Potentially, this will not only improve in- tates the assembly and use of open-electronics instru- stitutes’ international visibility and reputation but may ments, but will help ensure that most maintenance, re- also help in acquiring public funding. pairs, and upgrades can be performed in-house quickly, by the users themselves, with minimal fabrication ex- Benefits to the scientific community and penses beyond parts. Such an approach provides the op- funders portunity to practice a more sustainable small-scale cir- cular economy concept at institutions (Prendeville et al. Improved transparency and reproducibility 2016) because of the ability to replace or update compo- Transparency and reproducibility are hallmarks of the nents of the hardware as opposed to disposing of entire scientific method, but high costs and lack of documen- units of outdated proprietary instruments. Such control tation of procedures and tools in published methods over sustaining critical scientific equipment is crucial commonly prevent effective replication (Baker 2016). for all research institutions, but especially so for insti- Open electronics offer an opportunity to counter this tutes with scarce funding or in countries where local issue. Published applications based on open electronics technical support from specialized vendors lacks or is become technically and financially easier to reproduce prohibitively expensive. through decreased reliance on proprietary solutions. At While small custom setups are most common among the same time, as more new developments about open open-electronics projects, they are by no means lim- electronics are published, the more it will become ac- ited to these. For example, in order to grow and main- cepted and established to transparently communicate tain their large infrastructure sustainably, the European detailed blueprints of the solutions employed. Center for Nuclear Research (CERN) built the elec- tronic components of their particle accelerator with Decoupling of research capability from funding open-source hardware (van der Bij et al. 2012). Follow- The user-centered and iterative nature of open electron- ing this uncommon path, they have been simultane- ics better facilitates and enables specialized research ously innovating in electronics CAD software KiCAD than most proprietary solutions. For example, the use of and the CERN Open Hardware License (Svorc and Katz electronics in biological research in harsh ecosystems, 2019) as well as including openness as a requirement of such as wave-swept rocky shores or remote deserts, is their manufacturing contracts, thereby avoiding vendor difficult, and equipment that is not specifically designed lock-in. One resource example which was developed in for it may easily become damaged or lost. In this con- this context but is now used across academia and in- text, researchers either secure more funds to cover the dustry, is the White Rabbit, the current gold standard to losses of expensive material or down-scale the research achieve ultra-fast data transfer synchronization in Eth- line. Alternatively, open electronics can be efficiently ernet networks (Moreira et al. 2009). harnessed to develop cheaper and fully fit-for-purpose 1068 M. Oellermann et al. equipment (Gandra et al. 2015), while minimizing the lance and home automation) into scientific experiments cost incurred when losses occur. These and equivalent (Jolles 2021a), it also offers great opportunity to facili- solutions alleviate the entry cost of many research top- tate bidirectional collaborations with the public and sci- ics and contribute to a greater decoupling of research ence. Funders and society increasingly expect scientists capability from funding, ultimately facilitating the ex- to engage more actively with the public to improve the ploration of novel research lines and supporting investi- uptake and application of scientific knowledge (Hunter gations of early career researchers and scientists world- 2016). At the same time, the public increasingly de- wide, which have reduced access to infrastructure and mands to actively engage in the scientific process, to funding. an extent that citizens partner or even co-author with professional scientists (Breen et al. 2015; Mazumdar High innovation and collaboration potential et al. 2017). However, access to scientific instruments It is a common prejudice that open-source develop- has partly hampered such engagements and bottom-up ment conflicts with commercialization and industry initiatives by non-scientists to develop scientific ques- collaboration. Just like successful open-source software tions themselves (Mazumdar et al. 2017; Ostermann- companies, open electronics are an excellent basis for Miyashita et al. 2021). Open electronics can overcome commercial knowledge transfer. Well-designed scien- this barrier by providing cost-effective and interactive tific instruments mutually involve developers (typically tools that can be easily rebuilt by non-experts while hav- engineering-oriented teams but increasingly also open ing the capacity to generate high-quality scientific data electronics “makers”) and end-users (typically non- (Weeser et al. 2018). The user-centered, iterative, and engineering-oriented researchers) during the innova- modular design that is inherent to open-electronics so- tion and development process. However, often, such lutions further enhances a smooth exchange of knowl- user-centered design is not achieved due to the lack or edge and technical solutions between professional and ineffective communication between both groups. Open community scientists. Established design-sharing plat- electronics can overcome this by enabling end-users to forms such as Thingiverse enable direct feedback from become innovators and raising researchers’ basic elec- public communities and potentially accelerate the evo- tronics and programming skills to ease communica- lution of scientific open-electronics equipment. Thus, tion and collaboration between developers and scien- open electronics are well suited to make science broadly tific users. reproducible and more accessible for new collaboration Such additional technical skills by end-users en- opportunities. sure a better grasp of current technological bound- aries, permitting the establishment of goals that are si- multaneously realistic and ambitious. This can even Barriers and trade-offs lead to new commercial products. For example, some To reap the remarkable potential of open electronics now-commercial hardware options in the field of ecol- for biological research and science in general, signifi- ogy came from in-house custom laboratory solutions cant educational, collaborative, and technical barriers (e.g., Audiomoth [Open Acoustic Devices, Southamp- need to be overcome. For example, while the Raspberry ton, UK], ElectricBlue [ElectricBlue, Porto, Portugal]). Pi has large application potential, most researchers still At the same time, this user-centered and iterative de- lack basic awareness of such devices and their capac- velopment approach speeds up development cycles, ities and hence their uptake remains limited (Jolles which often results in fully functional solutions, and 2021a). One major reason is the lack or insufficient in many cases, it further enhanced by free user con- documentation of open-electronics setups in scientific tributions. Those and further advantages (e.g., fast- publications, which confines its visibility and the for- adaptation, easy user engagement, and advertisement) mation of any substantial academic “Maker” commu- can outweigh the disadvantages of such open-source nity (Glenn and Alfredo 2010; Harnett 2011). Instead, business models (e.g., reduced profit timeframe after in- many open-electronics techniques are spread among novation cycles are stopped, less acceptance of excessive collaborators in an informal fashion with limited reach. price margins) and provide rewarding opportunities for However, initiatives exist that aim to increase the vis- commercial developers and scientific users alike (Pearce ibility of open hardware solutions, such as the Open 2017). Neuroscience network (Maia Chagas 2021), the Open Source Ecology, Wildlab.net, and Conservation X Labs Bidirectional knowledge transfer between public and science (see Table 2), or new journals documenting open hard- While an increasing number of scientists have been in- ware designs in a systematic fashion, like the Jour- spired to integrate many of the openly shared electron- nal of Open Hardware (Murillo and Wenzel 2017) and ics solutions (e.g., home applications such as surveil- HardwareX. Open hardware in science: the benefits of open electronics 1069 Further, open electronics are poorly visible in disci- to increase. Much research requires highly customized plines where they do not define the research goal. For solutions without providing the quantitative scale for example, a Web of Science search showed that engi- commercial vendors to manufacture low-priced op- neers dominate the use of open-electronics boards listed tions. Thus, researchers can choose to purchase costly in Fig. 3 (36.1%), followed by Computer Science (26.4%, niche products, contract companies to develop and re- Fig. 1B, Supplementary Information File S1). However, pair customized applications at significant costs, or ac- a more detailed full-text analysis showed that although cess other available resources such as an institute’s work- biological sciences dominate the use of these boards shop and engineering staff. While the latter two can be within PLOS journals (34%, n = 86), before engineering time-consuming too, many researchers lack access to (11%, n = 27), only 5% of these articles reported their the necessary funding, local specialized companies, or use in the abstract (for detailed analysis, see Supplemen- infrastructure support. For them, open electronics may tary Information File S2). In addition, authors do not al- be the only pathway to execute the intended research ways mention clearly if and how they applied open elec- with time investment being a relatively minor trade-off. tronics in their research, making it difficult to identify Local peer-knowledge can be key to shorten time its actual use across subject areas. With improved ac- investments for learning and development of open- knowledgment, researchers will recognize its value at a electronics setups, particularly when complexity in- broader scale and potentially generate more associated creases. However, skilled users within departments or research, innovation, and public interest. institutions are still rare and collaborations with en- Another barrier is the fragmentation of the existing gineering scientists unlikely due to the lack of shared open-electronics user community within institutions goals or benefits. Also, locally established knowledge and across countries and subject domains, hindering may not sustain if scientists leave institutions. In con- the exchange and consolidation of knowledge (Fig. 1A trast, companies normally provide good product doc- and B). This is because within institutions and depart- umentation and customer support, yet more complex ments, there is often little support infrastructure for instrumentations require intense training at additional development, educational resources, and community- costs, and product support may cease for old products building, such as user-run “Maker” workshops (Maia or when companies cease to exist. To improve the ac- Chagas 2018). Across countries, peer-to-peer knowl- cess and sustainability of local knowledge, good docu- edge exchange is largely limited to existing collabora- mentation, the establishment of local open-electronics tion networks, often biased by geographical distance or user networks, and a structured knowledge transfer will socio-cultural cohesion (Hennemann et al. 2012), al- be key. This will also support self-repair and mainte- though open electronics’ scientific publication output, nance of open-electronics equipment by users not in- led by and linked between China, the United States, volved in their development. In comparison, repair of and India, indicates promising shifts to break those tra- proprietary products may become expensive once war- ditional patterns (Fig. 1A, Supplementary Information ranty expires and self-repair impossible due to sealed File S1). instruments, inaccessible hardware documentation, or Other important trade-offs of open electronics are surface mount electronics. Finally, while broad expe- time investment to learn and prototype, difficulties to rience with and detailed knowledge of coding is still access or sustain expertize and knowledge, or the sus- a common perceived barrier for people to get started tained ability to operate and maintain equipment. These with open electronics, for most projects, this is actually need to be weighed against the benefits and trade-offs not required. Furthermore, researchers can rely on easy- posed by proprietary alternatives, researchers’ needs to to-use libraries to work with sensors and actuators and access customized solutions, and their individual time numerous online resources that provide tutorials with or financial limitations. For example, if there are afford- ready-to-paste code (Table 2), thereby significantly low- able proprietary solutions that fulfill the research need ering the skill requirements for non-programmers. The (e.g., temperature measurement), then open electronics discussed trade-offs may weigh differently depending may not provide substantial benefits. Also, more com- on the individual knowledge access and research sup- plex setups may require significant time investments port and may decrease with time as skills and collabo- for optimization, repair, and maintenance, particularly rations increase. during early development stages, which may outweigh We believe that over time many of the discussed bar- their initial lower costs compared to well-tested and es- riers and trade-offs can be offset, beginning by increas- tablished commercial alternatives. However, once the ing the visibility of the tools themselves and a cultivation need for customization increases (e.g., additional sen- of an institutional and global collaborative community sors with cloud data transmission), the access to suitable around their use, to build confidence, electronics liter- commercial alternatives declines and their price tends acy, and local knowledge pools. Global networks such 1070 M. Oellermann et al. Figure 2. Diagrammatic representation of the potential steps for incorporating open electronics into one’s research. It is best to begin with a starter kit to explore its potential. Tutorials are useful for building initial skills, such as to set up a sensor to measure the temperature of an aquarium. Delving further into the many (online) resources available, basic systems can be expanded to perform more advanced tasks, such as plotting the temperature data in real time on a simple website and sending warning emails whenever values cross thresholds. Such a system can then be easily and affordably replicated and shared with the broader community. as the Gathering for Open Science Hardware (Murillo Implement and iterate to advanced et al. 2018) and Open-Source Ecology (see Table 2), and applications an increasing number of scientific societies hosting ded- Once researchers obtained their first electronic and icated symposia (e.g., the Annual Meeting of the Soci- programming skills, open electronics can be employed ety for Experimental Biology or the Society for Integra- in simple biological experiments. This can include tive and Comparative Biology), are an excellent start to field logging of temperature or humidity (Baker 2014), spread awareness and share innovative open-electronics recording images or videos in experiments (Jolles 2020), solutions across disciplines and budgets. or detecting wildlife using weatherproof camera traps (Droissart et al. 2021). At this level already, repro- Beginners toolbox ducibility and scalability are important to consider, so that initial designs can be expanded to increase Starting simple complexity, throughput, and replication by other re- Once the decision has been made to use open elec- searchers. Simple designs help to easily replicate pro- tronics, the best way to begin is to start creating totypes, reduce time for troubleshooting during oper- simple set-ups, to gain experience in how open ation, and lower reproduction costs. For example, this electronics work, and to inspire first custom appli- can lead to a gradual expansion, first using one cam- cations (Fig. 2). Hobby electronics starter kits (e.g., era system to phenotype plants (Tovar et al. 2018) up to www.sparkfun.com, www.adafruit.com) provide all large-scale and simultaneous high-throughput pheno- essential electronic components, such as breadboard, typing of 1800 plants using 180 Raspberry Pis and cam- cables, resistors, or LEDs, and introduce researchers eras (Tausen et al. 2020). Such a stepwise expansion is to the large range of sensors and actuators useful in advisable and may require some iterations of replace- biological research (Table 1). These kits come with ment or optimization of electronic components and micro-controllers and single-board computers, for programming code, for which fellow users may provide which a large range of options exist (see Fig. 3 for support in the numerous online forums (Table 2). Apart an overview of devices). While the micro-controllers from scale, applications may also grow in complexity, as are mostly constrained to one programming language most open-electronics boards allow the connection of (e.g., C/C++, micropython), single-board computers multiple and different types of sensors, cameras, and ac- offer more programming choices (e.g., Python, C++, tuators, either directly or via easy-to-connect and easy- JavaScript etc.). For beginners especially, Arduino to-read breakout boards. For example, this enables re- and Raspberry Pi are recommended as they have the searchers to measure multiple variables simultaneously most documentation and support available (see Table in single setups such as moisture, temperature, or dif- 2). One can then gradually progress to other more ferent types of gases in soil (Bitella et al. 2014; Dorji specialized applications and boards (Fig. 3), using et al. 2017); remotely analyze water bodies and land- the many additional introductory learning resources scapes using multispectral sensors carried by drones (Table 2). (Bokolonga et al. 2016); or monitor the bio-acoustic en- Open hardware in science: the benefits of open electronics 1071 Figure 3. Overview of some of the key micro-controllers (MC) and single-board computers (SBC) on the market, designed to enable the creation of open-electronics applications by non-experts. Models are rated for their price, skill level, performance, flexibility, resources, and support available, and possibility to run machine learning applications, based on publicly available model specifications (dimensions, RAM, CPU, number of cores, GPU, power consumption, number of interfaces, wired and wireless connectivity, video recording capability, storage), popularity, and authors’ experiences. Manufacturers: 1 Adafruit Industries, New York, USA; 2 Arduino, New York, USA; 3 ASUSTek Computer Inc., Taipei, Taiwan; 4 Texas Instruments Incorporated, Dallas, USA; 5 Nvidia Corporation, Santa Clara, USA; 6 Espressif Systems, Shanghai, China; 7 Hardkernel Co., Ltd., GyeongGi, South Korea; 8 Raspberry Pi Foundation, Cambridge, UK; 9 Red Pitaya, Solkan, Slovenia; 10 Udoo, Arezzo, Italy. vironment in tropical or temperate habitats (Whytock et erly acknowledge the use of open electronics and share al. 2016). A large variety of C or Python libraries enable their designs, methodologies, and knowledge with the the increase of application complexity beyond hardware broader community. This can be best achieved by without professional programming skills to, for exam- following a three-step standard practice to acknowl- ple, send automatic email notifications, share and visu- edge open-electronics solutions in research publica- alize data life, program custom GUIs to control elec- tions (Fig. 4). tronics and devices, or analyze “data on board,” such as the simultaneous behavioral monitoring of up to 1400 1) The use of “open electronics” or “open hardware” flies, using machine learning algorithms (Geissmann et should be clearly highlighted and mentioned using al. 2017) or constructing detailed 3D models of arthro- standard terms in the abstract and keywords. This pods (Plum and Labonte 2021). Such complex appli- enables users to easily locate relevant publications in cations may seem daunting for beginners, but an in- public databases and metadata searches. Due to the creasing number of traditional as well as dedicated large diversity of boards and new developments, sim- open hardware scientific journals (e.g., Journal of Open ply using, for example, micro-controller names will Hardware or HardwareX) offer detailed construction not be comprehensive enough to locate all relevant and coding instructions to ease reproduction. sources. The introduction too should contain a short explanation of the open-electronics application. 2) The methods section should contain detailed infor- Acknowledge and share open electronics mation describing the open-electronics application. For beginners and advanced users alike, good docu- Those details should enable to fully replicate the mentation is key to enable others to reproduce setups setup and therefore include a list of all components and further improve and adapt open-electronics de- with their model, supplier, and price; written or illus- signs. Therefore, it is essential that researchers prop- trated fabrication instructions; and photos, illustra- 1072 M. Oellermann et al. Figure 4. Concept diagram illustrating a three-step standard practice to best acknowledge the use of open electronics in science. Vector graphic courtesy of Arto Lereh Saraga, Jaohuarye, Uizin, and Adrien Coquet from NounProject.com. tions, 3D files, or videos (e.g., Eggert et al. 2020), and ways. With the increasing capabilities of electronic com- computer aided design (CAD) files where applicable. ponents and sensors, and computers becoming more If methods are not the focus of the study and space is powerful at decreasing size and cost, open electronics limited, then this section should include minimum a are likely to become progressively used and integrated list of all components used with the exact model and in our day-to-day life, and over time become a standard supplier and a reference to the detailed constructions component of the scientific toolbox. This, in turn, will blueprint in the supplements or external sources. result in new and cutting-edge technologies to be im- 3) In addition to providing those details in the meth- plemented at a broader scale, and help tech-innovation ods section or supplements, one should publish and to expand to other non-engineering disciplines. share projects on online platforms such as GitHub An important step to increase the uptake of open using Markdown files, on institutional or public electronics is improved support by funding organiza- Wikis, or even create a dedicated website using free tions, such as to grant researchers dedicated time to de- services such as GitHub pages or WordPress. This velop, build, publish open-electronics applications, and has the benefit of receiving direct feedback from request open hardware alternatives in compulsory in- other users to improve and further develop the ap- strument bids. Furthermore, institutions can foster lo- plication. To ensure reproducibility of the published cal “ScienceMaker” communities, by providing institu- version, the exact version of this detailed documen- tional “MakerHubs” or workshops, where researchers tation should be archived on a scientific webpage ca- can prototype and exchange knowledge and ideas with pable of generating a digital object identifier (DOI), others, as well as include electronics and programming such as Zenodo.org and this link should be shared in training to the institutional career development portfo- the publication itself. Depending on its novelty, one lio. And scientific communities can start or join open may also decide to write up a methods paper about hardware initiatives, for example, Global Open Science the specific device and its applications, such as in Hardware community (Murillo et al. 2018); organize the mentioned open hardware journals. This has the dedicated conferences, sessions, or workshops to form added benefit of creating a citable reference, which networks; create standards; and foster open electronics can improve publication output and researcher’s vis- across disciplines (Bonvoisin et al. 2020). ibility outside their own discipline, as well as serve as In this paper, we presented the multi-facetted ben- a proof of prior art of intellectual property. efits open electronics may offer to researchers, insti- tutions, and the scientific community, and highlighted their utility and potential for biological research and sci- Outlook ence as a whole, while also noting important barriers The potential of open-electronics applications is endless and trade-offs, and avenues to overcome those, includ- and can greatly benefit individual researchers, institu- ing a beginner’s guide. We hope our review will help tions, and the scientific community in a broad variety of foster a broader awareness and uptake of open electron- Open hardware in science: the benefits of open electronics 1073 ics across the biological sciences, and from research to ference on Advanced Communication Technologies and Net- education and outreach, and thereby to ultimately help working (CommNet), Marrakech, Morocco. increase the innovation, reproducibility, and democra- Allan BM, Nimmo DG, Ierodiaconou D, VanDerWal J, Koh LP, Ritchie EG. 2018. Futurecasting ecological research: the rise of tization of science. technoecology. Ecosphere 9:e02163. Arunachalam A, Andreasson H. 2021. RaspberryPi-Arduino Supplementary data (RPA) powered smart mirrored and reconfigurable IoT facility for plant science research. Internet Technol Lett.5: e272. Supplementary Data available at ICB online. Baker E. 2014. Open source data logger for low-cost environmen- tal monitoring. Biodivers Data J 2:e1059. Author contributions Baker M. 2016. 1,500 scientists lift the lid on reproducibility. Na- ture 533:452–4. Tabular author contributions modified after the CRediT Beckler J, Gray K, Carothers B, Fields H, Currier B, Schloesser R. system. All authors contributed critically to the drafts 2018. Engaging high school students and teachers through an and gave final approval for publication. ocean-observing technology STEM outreach club. Current: J Mar Educ 31:13–21. Beddows PA, Mallon EK. 2018. Cave pearl data logger: a flexible arduino-based logging platform for long-term monitoring in harsh environments. Sensors 18:530. Bitella G, Rossi R, Bochicchio R, Perniola M, Amato M. 2014. A novel low-cost open-hardware platform for monitoring soil water content and multiple soil-air-vegetation parameters. Sensors 14:19639–59. Bokolonga E, Hauhana M, Rollings N, Aitchison D, Assaf MH, Das SR, Biswas SN, Groza V, Petriu EM. 2016 A compact mul- tispectral image capture unit for deployment on drones. IEEE International Instrumentation and Measurement Technology Conference Proceedings, 2016, Taipei, Taiwan. Bonvoisin J, Molloy J, Häuer M, Wenzel T. 2020. Standardisation of practices in open source hardware. J Open Hardw 4:2. Boudoire G, Liuzzo M, Cappuzzo S, Giuffrida G, Cosenza P, Derrien A, Falcone EE. 2020. The SoilExp software: an open- source Graphical User Interface (GUI) for post-processing Acknowledgments spatial and temporal soil surveys. Comput Geosci 142:104553. We would like to thank Jenny Molloy and Andre Maia Boulton G, Campbell P, Collins B, Elias P, Hall W, Laurie G, O’Neill O, Rawlins M, Thornton J, Vallance P. 2012. Sci- Chagas whose comments helped improve and clarify ence as an open enterprise. London, UK: The Royal Society. the manuscript. p. 1–104. Breen J, Dosemagen S, Warren J, Lippincott M. 2015. Mapping Funding grassroots: geodata and the structure of community-led open environmental science. ACME: Int J Crit Geogr 14:849–73. M.O. was supported by a German Research Foun- Bridge ES, Wilhelm J, Pandit MM, Moreno A, Curry CM, Pear- dation fellowship OE 658/2. R.L.T. was supported son TD, Proppe DS, Holwerda C, Eadie JM, Stair TF et al. 2019. by a California Sea Grant Delta Science Postdoc- An Arduino-based RFID platform for animal research. Front toral Fellowship SWC 20–42 Project R/SF-107. Ecol Evol 7:257. T.W. was supported by the EMBL, the Fellowship Colavizza G, Hrynaszkiewicz I, Staden I, Whitaker K, Freies Wissen by Wikimedia DE and co, and by McGillivray B. 2020. The citation advantage of linking publications to research data. PLoS One 15:e0230416. ANID/FONDECYT/INICIACION/N◦ 11200666. R.S. Costa DG, Duran-Faundez C. 2018. Open-source electronics was funded by FEDER through POR Norte (NORTE- platforms as enabling technologies for smart cities: recent de- 01–0145-FEDER-031053); and by National Funds velopments and perspectives. Electronics 7:404. through FCT (PTDC/BIA-BMA/31053/2017 and Divani D, Patil P, Punjabi SK. 2016 Automated plant water- CEECIND/01424/2017). D.O. was supported by INTA ing system. 2016 International Conference on Computation of through projects 2019-PE-E6-I128-001, 2019-PE-E6- Power, Energy Information and Commuincation (ICCPEIC), I114-001, and REC I117. Melmaruvathur, India. Dorji U, Pobkrut T, Siyang S, Chaiyasit T, Choden P, Kerd- charoen T. 2017 Development of real time smart elec- References tronic nose system for monitoring organic abundance in soil. Abdelouahid RA, Debauche O, Mahmoudi S, Marzak A, Man- 2017 14th International Conference on Electrical Engineer- neback P, Lebeau F. 2020. Smart Nest Box: IoT based nest ing/Electronics, Computer, Telecommunications and Infor- monitoring in artificial cavities. 2020 3rd International Con- mation Technology (ECTI-CON), Phuket, Thailand. 1074 M. Oellermann et al. Droissart V, Azandi L, Onguene ER, Savignac M, Smith TB, De- Jolles JW. 2021a. Broad-scale applications of the Raspberry Pi: blauwe V. 2021. PICT: a low-cost, modular, open-source cam- a review and guide for biologists. Methods Ecol Evol 12: era trap system to study plant–insect interactions. Methods 1562–79. Ecol Evol 12:1389–96. Jolles JW. 2021b The Raspberry Pi guide for scientists and anyone Eggert S, Mieszczanek P, Meinert C, Hutmacher DW. 2020. else!. https://raspberrypi-guide.github.io/ (1 March 2022, date OpenWworkstation: a modular open-source technology for last accessed). automated in vitro workflows. HardwareX 8:e00152. Lai J, Lortie CJ, Muenchen RA, Yang J, Ma K. 2019. Evaluating Foley CJ, Sillero-Zubiri C. 2020. Open-source, low-cost modu- the popularity of R in ecology. Ecosphere 10:7. lar GPS collars for monitoring and tracking wildlife. Methods Lertvilai P, Durand P. 2020. The in situ Plankton Assemblage eX- Ecol Evol 11:553–8. plorer (IPAX): an inexpensive underwater imaging system for Gandra M, Seabra R, Lima FP. 2015. A low-cost, versatile data zooplankton study. Methods Ecol Evol 11:1042–8. logging system for ecological applications. Limnol Oceanogr Lewinski M, Bramkamp Y, Koster T, Staiger D. 2020. SEQing: Methods 13:e10012. web-based visualization of iCLIP and RNA-seq data in an in- Geissmann Q, Garcia Rodriguez L, Beckwith EJ, French AS, Ja- teractive python framework. BMC Bioinf 21:113. masb AR, Gilestro GF. 2017. Ethoscopes: an open platform for Maia Chagas A. 2018. Haves and have nots must find a better way: high-throughput ethomics. PLoS Biol 15:e2003026. the case for open scientific hardware. PLoS Biol 16:8. Glenn M, Alfredo H. 2010. IEEE humanitarian projects: open Maia Chagas A. 2021. Open Neuroscience. https://open-neurosc hardware for the benefit of the poorest nations. Open Source ience.com (1 March 2022, date last accessed). Business Resource (December 2010). https://timreview.ca/art Mazumdar S, Wrigley S, Ciravegna F. 2017. Citizen science and icle/401 (1 March 2022, date last accessed). crowdsourcing for Earth observations: an analysis of stake- Griffiths AGF, Kemp KM, Matthews K, Garrett JK, Griffiths DJ. holder opinions on the present and future. Remote Sens 2017. Sonic Kayaks: Environmental monitoring and experi- 9:87. mental music by citizens. PLoS Biol 15:e2004044. McCray AT, Berman F, Carroll M, Ginther D, Miller R, Schiffer Gurdita A, Vovko H, Ungrin M. 2016. A simple and low-cost P, Seidel E, Szalay A, Tauxe L, Xu H. 2018. Open science by monitoring system to investigate environmental conditions in design: realizing a vision for 21st century research. Washing- a biological research laboratory. PLoS One 11:e0147140. ton DC: The National Academies Press. https://doi.org/10.172 Harnett C. 2011. Open source hardware for instrumentation and 26/25116. measurement. IEEE Instrum Meas Mag 14:34–8. Moreira P, Serrano J, Wlostowski T, Loschmidt P, Gaderer G. Hasan M, Biswas P, MD Bilash TI, MD Dipto AZ.2018 Smart 2009 White rabbit: sub-nanosecond timing distribution over home systems: overview and comparative analysis. 2018 ethernet. 2009 International Symposium on Precision Clock Fourth International Conference on Research in Compu- Synchronization for Measurement, Control and Communica- tational Intelligence and Communication Networks (ICR- tion, Brescia, Italy. CICN), Kolkata, India. Muenchen RA. 2012. The popularity of data analysis software. Hennemann S, Rybski D, Liefner I. 2012. The myth of global sci- Data Science Software Reviews. r4stats.com/articles/popularit ence collaboration—collaboration patterns in epistemic com- y/ (1 March 2022, date last accessed). munities. J Informetr 6:217–25. Murillo LFR Wenzel T 2017. Welcome to the Journal of Open Hohlbein J, Diederich B, Marsikova B, Reynaud EG, Holden S, Hardware. Journal of Open Hardware 1:1 Jahr W, Haase R, Prakash K. 2021. Open microscopy in the Murillo LFR, Molloy J, Dosemagen S, Build D, Grey F, Irujo life sciences: Quo Vadis? arXiv:2110.13951v1. GP, Toivonen H, Serrano J,Maestre JP, Bernaldo P et al. 2018. Hunter P. 2016. The communications gap between scientists and Global open science hardware roadmap. https://openhardware public. EMBO Rep 17:1513–5. .science/global-open-science-hardware-roadmap/ (1 March Ibrahim M, Elgamri A, Babiker S, Mohamed A. 2015 Inter- 2022, date last accessed). net of things based smart environmental monitoring using Open Neuroscience. 2021. Open Neuroscience. https://open-neu the Raspberry-Pi computer. 2015 Fifth International Confer- roscience.com/. (last accessed 1 August 2012) ence on Digital Information Processing and Communications Ostermann-Miyashita EF, Pernat N, König HJ. 2021. Citizen sci- (ICDIPC), Sierre, Switzerland. ence as a bottom-up approach to address human–wildlife con- Jiang Q, Kresin F, Bregt AK, Kooistra L, Pareschi E, van flicts: from theories and methods to practical implications. Putten E, Volten H, Wesseling J. 2016. Citizen sensing for Conserv Sci Pract 3:e385. improved urban environmental monitoring. J Sens 2016: Pearce JM. 2012. Building research equipment with free, open- 10. source hardware. Science 337:1303–4. Johnson BI, De Moraes CM, Mescher MC, Firn J. 2016. Manip- Pearce JM. 2013. Open-source lab: how to build your own hard- ulation of light spectral quality disrupts host location and at- ware and reduce research costs. Oxford: Elsevier. tachment by parasitic plants in the genus Cuscuta. J Appl Ecol Pearce JM. 2015. Quantifying the value of open source hardware 53:794–803. development. Mod Econ 6:1–11. Jolles JW, Boogert NJ, Sridhar VH, Couzin ID, Manica A. Pearce JM. 2016. Return on investment for open source scientific 2017. Consistent individual differences drive collective behav- hardware development. Sci Public Policy 43:192–5. ior and group functioning of schooling fish. Curr Biol 27: Pearce JM. 2017. Emerging business models for open source 2862–8. hardware. J Open Hardw 1:2. Jolles JW. 2020. pirecorder: Controlled and automated image and Plum F, Labonte D. 2021. scAnt-an open-source platform for the video recording with the raspberry pi. J Open Source Softw creation of 3D models of arthropods (and other small objects). 5:2584. PeerJ 9:e11155. Open hardware in science: the benefits of open electronics 1075 Poongothai M, Subramanian PM, Rajeswari A. 2018 Design and Tadres D, Louis M. 2020. PiVR: an affordable and versatile implementation of IoT based smart laboratory. 2018 5th Inter- closed-loop platform to study unrestrained sensorimotor be- national Conference on Industrial Engineering and Applica- havior. PLoS Biol 18:25. tions (ICIEA), Singapore. Tausen M, Clausen M, Moeskjær S, Shihavuddin A, Dahl AB, Prendeville S, Hartung G, Purvis E, Brass C, Hall A. 2016 Janss L, Andersen SU. 2020. Greenotyper: image-based plant Makespaces: from redistributed manufacturing to a circular phenotyping using distributed computing and deep learning. economy. In: Setchi R, Howlett R, Liu Y, Theobald P (eds.) Front Plant Sci 11:1181. Sustainable design and manufacturing 2016, Vol. 52, Cham: Tovar JC, Hoyer JS, Lin A, Tielking A, Callen ST, Elizabeth Springer International Publishing. p. 577–88. Castillo S, Miller M, Tessman M, Fahlgren N, Carrington JC R Core Team. 2021 R: A language and environment for statistical et al. 2018. Raspberry Pi-powered imaging for plant pheno- computing, R Foundation for Statistical Computing. Vienna. typing. Appl Plant Sci 6:e1031. https://www.R-project.org/ (1 March 2022, date last accessed). Trasviña-Moreno CA, Blasco R, Marco Á, Casas R, Trasviña- RaspberryPiFoundation. 2020. Annual review for the Raspberry Castro A. 2017. Unmanned aerial vehicle based wireless sensor Pi foundation. https://www.raspberrypi.com (1 March 2022, network for marine-coastal environment monitoring. Sensors date last accessed). 17:460. Roe P, Ferroudj M, Towsey M, Schwarzkopf L. 2018 Catching van der Bij E, Serrano J, Wlostowski T, Cattin M, Gousiou toad calls in the cloud: commodity edge computing for flexible E, Sanchez PA, Boccardi A, Voumard N, Penacoba G. 2012. analysis of big sound data.2018 IEEE 14th International Con- Open hardware for cern’s accelerator control systems. J In- ference on e-Science (e-Science), Amsterdam, Netherlands. strum 7:C01032. Schwaner MJ, Freymiller GA, Clark RW, McGowan CP. 2021. van Helden P. 2012. The cost of research in developing countries. How to stick the landing: kangaroo rats use their tails to reori- EMBO Rep 13:395. ent during evasive jumps away from predators. Integr Comp Vicente-Saez R, Martinez-Fuentes C. 2018. Open science now: a Biol 61:442–54. systematic literature review for an integrated definition. J Bus Sethi SS, Ewers RM, Jones NS, Orme CDL, Picinali L. 2018. Res 88:428–36. Robust, real-time and autonomous monitoring of ecosystems Virag D, Homolak J, Kodvanj I, Knezovic A, Osmanovic Barilar J, with an open, low-cost, networked device. Methods Ecol Evol Salkovic-Petrisic M. 2021. Repurposing a digital kitchen scale 9:2383–7. for neuroscience research: a complete hardware and software Setyowati V, Muninggar J, Shanti.N.A. MRS. 2017. Design of cookbook for pasta. Sci Rep 11:8. heart rate monitor based on piezoelectric sensor using an Ar- Von Hippel E. 2005. Open source software projects as user inno- duino. J Phys Conf Ser 795:012016. vation networks. Persp free open source softw 267–78. Shipley JR, Kapoor J, Dreelin RA, Winkler DW, Lecomte N. 2017. Von Hippel E. 2007. Horizontal innovation networks—by and for An open-source sensor-logger for recording vertical move- users. Ind corp chang 16:293–315. ment in free-living organisms. Methods Ecol Evol 9:465–71. Weeser B, Stenfert Kroese J, Jacobs SR, Njue N, Kemboi Z, Ran Siregar B, Efendi S, Pranoto H, Ginting R, Andayani U, Fahmi F. A, Rufino MC, Breuer L. 2018. Citizen science pioneers in 2017. Remote monitoring system for hydroponic planting me- Kenya—a crowdsourced approach for hydrological monitor- dia. 2017 International Conference on ICT For Smart Society ing. Sci Total Environ 631-632:1590–9. (ICISS), Tangerang, Indonesia. Whytock RC, Christie J, Jones K. 2016. Solo: an open source, cus- Steffens S, Nüßer L, Seiler TB, Ruchter N, Schumann M, Döring tomizable and inexpensive audio recorder for bioacoustic re- R, Brinkmann M. 2017. A versatile and low-cost open source search. Methods Ecol Evol 8:308–12. pipetting robot for automation of toxicological and ecotoxico- Wilkinson M, Bell MC, Morison JIL, Alberto Silva C. 2021. logical bioassays. PLoS One 12. A Raspberry Pi-based camera system and image pro- Stephenson W, Donlin LT, Butler A, Rozo C, Bracken B, Rashid- cessing procedure for low cost and long-term monitor- farrokhi A, Goodman SM, Ivashkiv LB, Bykerk VP, Orange DE ing of forest canopy dynamics. Methods Ecol Evol 12: et al. 2018. Single-cell RNA-seq of rheumatoid arthritis syn- 1316–22. ovial tissue using low-cost microfluidic instrumentation. Nat Wong BG, Mancuso CP, Kiriakov S, Bashor CJ, Khalil AS. 2018. Commun 9:791. Precise, automated control of conditions for high-throughput Stewart FJ, Kinoshita M, Arikawa K. 2017. A novel display sys- growth of yeast and bacteria with evolver. Nat Biotechnol tem reveals anisotropic polarization perception in the motion 36:614–23. vision of the butterfly Papilio xuthus. Integr Comp Biol 57: Youngblood M. 2020. A Raspberry Pi-based, RFID-equipped 1130–8. birdfeeder for the remote monitoring of wild bird populations. Svorc J, Katz A. 2019. Breathe in, breathe out: how open hard- Ringing Migr 34:25–32. ware licensing can help save the world. J Open Law Technol Zamore SA, Araujo N, Socha JJ. 2020. Visual acuity in the flying Soc 11:49–56. snake, Chrysopelea paradisi. Integr Comp Biol icaa143:1–14.