Prelude
Contents
Prelude#
Gathering#
In 2019, Eirini Malliaraki curated a collection of open data in the field of environmental information. Data for Environmental Intelligence gives an overview of open and freely available resources on the state of our natural environment. It was created with the aim of helping developers, journalists, scientists – ultimately, anyone with access to an internet connection – to find information on how our planet’s natural systems are changing.
In the same year, Tobias Augspurger curated a list of open source robotic tools for professional robotic development with the Robotic Operation System (ROS). Based on his experience leading the development of a complete robotic software stack at DHL, primarily based on open source tools, he became convinced that open source will play an increasingly important role in the co-evolution of technology and society.
Together with the online community protontypes, they launched the Open Sustainable Technology project to map all open source software repositories in the field of environmental sustainability. Their goal was to capture a complete picture of this emerging open source ecosystem. It quickly became clear that a wealth of information could be derived from this database that warranted further investigation.
In September 2021, Eirini and Tobias joined forces towards a common goal: to explore and understand the intersection of open source, environmental sustainability, and technology, and to share their findings and insights with the public. Further collaborations were agreed upon to build a global picture of the open source ecosystem. In 2022, Josh Hopkins joined to investigate open source projects in the context of digital public goods for communities and decision-makers, led by sustainability intelligence non-profit organisation Open Corridor.
Fig. 1 - Open source software and Open Data are the foundation of many modern innovation processes. This artwork is generated by the open source artificial intelligence Stable Diffusion.#
The Role of Open Source Software#
Open source relies on the openness of source code, the ability to modify it, and the freedom to distribute it. Open source software (OSS) is developed through the collaborative efforts of distributed communities, who work together to improve software solutions and share their modifications with others, resulting in an efficient allocation of resources and more robust and reliable products. It is foundational in establishing a digital commons, supporting open innovation communities, managing software development lifecycles, and informing social practices for managing shared resources.
While there is no central authority by design, the open source movement is guided by common principles, such as the Fundamental Principles, defined by the Open Source Initiative. At the same time, the Open Science movement and FAIR Principles have been instrumental in making scientific work transparent and collaborative.
Open source software has transformed industries, allowing for the rapid development of many of the advanced systems which our economies and societies rely upon. Today open source is dominant in 90% of cloud computing,1 82% of smartphones,2 97% of application development,3,4 and most artificial intelligence and data science.5 Linux is by far the most widely used operating system in embedded systems, internet infrastructure, and supercomputers, with all of the top 500 most powerful supercomputers in the world running a Linux distribution.6 Not surprisingly, 90% of the Fortune Global 500 leverage OSS.7. This trend is also evident in venture capital investments in commercial open source software with a total investment of $24B between 2020 and 2022.8 Several larger companies with annual revenues of more than $100M are on the rise.9 The total market for open source services is expected to reach $66.84B by 2026.10
Tip
Click the image to zoom into the dependency tree.
Fig. 2 - The dependency tree of PyBaMM, solving physics-based electrochemical DAE models by using state-of-the-art automatic differentiation and numerical solvers. The plot is created with pipdeptree
#
Without the high modularity made possible by open source, software development as we know it today would not exist. Package managers and indices such as PyPA and pip for Python enabled the simple and secure distribution of software components. This process of knowledge generation, adaptation and transformation has been known in mathematics and the natural sciences for centuries. Now it finds its application in open source software development. This approach is often described with the metaphor of “standing on the shoulders of giants”, which is explained in more detail in the chapter Open Sustainability Principles. Modern software development is based on the developments of thousands of small projects that have been released as open source over the last decades. Fig. 2 illustrates the connection of a typical open source project with several other releases within the ecosystem.
Besides core information technology infrastructure, open source software is becoming increasingly critical in environmental science,11 disaster impact assessment,12 energy efficiency,13 and sustainability14 in general. While technological advances are critical to addressing climate change and environmental sustainability, OSS used in the design, operation, and maintenance of human-engineered systems is rarely considered in its own right. This applies to technologies such as energy production and storage, as well as software that simulates and predicts complex natural systems including the atmosphere, biosphere, and hydrosphere.
Fig. 3 - The International Energy Agency is working with the National Renewable Energy Laboratory, Technical University of Denmark and the University of Maine to develop an open source 15-megawatt offshore reference wind turbine design with open simulation and development tools. License: Apache License 2.0#
Without observing and simulating vital natural systems, it is impossible to understand their interrelationships, or indicate how humans can preserve this unique, life-protecting world that has emerged in a completely hostile space. Mathematical models, technologies and measurement tools can provide us with independent reporting on the state of the planet, and the extent of human impacts. The knowledge accumulated over decades ensures that people around the world can understand how to preserve vital resources such as fresh water, fertile soil, clean air, and a stable climate. The in-depth understanding of the Earth and its natural, economic and social systems subsequently allows us to ask critical questions and make accurate predictions about human actions and associated impacts such as:
How does the Earth system respond to global anthropogenic changes?
What are the risks of exceeding Earth system thresholds? What are the remaining operating budgets? How do we set targets and allocate responsibility?
What quantity of greenhouse gases does a product and its supply chain release into the atmosphere?
How can the value of materials be retained and waste be reduced?
How does the demand for natural resources impact ecosystems, and how can biodiversity be conserved?
What are the most effective methods for capturing, storing, and distributing energy?
What risk does a particular technology pose to humans and the environment in the event of a failure?
Without open data, rigorous science and open software, it is impossible to make evidence-based and verifiable assessments about the feasibility of technologies and their potential impact. The manipulation and withholding of information related to the environmental impact of technologies and companies has a long history. From the fossil fuel industry withholding studies about climate change; to manipulation of data and measurements about emissions by the automotive industry; to a lack of communication of important environmental safety information – pollution and environmental disasters go hand in hand with opaque reporting, obfuscated data, and closed-sourced models. To avoid greenwashing and manipulation of environmental information, we need to ensure that claims about the sustainability of technology, the environmental footprints of products, and the impact of human actions are based on quantifiable and verifiable evidence.
A Brief History of Open Source Culture in Sustainability#
Open environmental data has a very long but not well-known history. Without the exchange of measurement data between meteorologists, it would have been impossible to achieve accurate weather forecasting in the early days of this scientific field. The origin of the World Meteorological Organization (WMO) stretches back almost 150 years, and international collaboration and data sharing have been central to its activities from the very beginning.
With the advent of the space age, the small number of weather satellites forced nations around the world to share data between different observatories in order to understand the state of the atmosphere and other Earth systems as a whole. The weather forecasts that became possible as a result of this innovation added significant value to all sectors of society. This open, planetary-scale computation project increased our awareness of anthropogenic climate change and our ability to act in this environment.
Despite increasingly strong commercial interests in civil satellite data, a strong culture of open data and open science has been established to date. 41% of the unclassified Earth observation satellite programs of the 10 largest nations today provide open data.15 Open access to this data has led to a rapid increase in the number of downloads shown in Fig. 4 using the example of the Landsat Archive. The creation of the Global Climate Observing System (GCOS), launched at the Second World Climate Conference, also committed parties to support international and intergovernmental sharing of data and analysis.16 The urgent need for a transition to a more sustainable society, and clear signs of anthropogenic climate change, have created multiple new movements and organisations – all with similar open ethos and mandates across domains:
In the energy sector, the OpenMod Initiative formed in 2014 with the goal of opening up energy models, so the sector as a whole can improve the quality, transparency, and credibility of its products and create better research and policy.
In 2017, academic Stefan Pfenninger wrote an article published in Nature urging scientists to “Free The Models” for the energy sector – a call to action supported by influential scientists like Auke Hoekstra and entrepreneurs like Michael Liebreich. This led to the sector pushing toward open data and open source code – a move that will be critical for the transition to a fossil fuel-free economy.
In 2018, Shuli Goodman founded the LF Energy, an open source organisation within the Linux Foundation that enables companies worldwide to develop energy systems related OSS tools collaboratively.
ClimateChange.ai convenes an open community and provides multiple educational resources around tackling climate change through artificial intelligence – another technological revolution that has been catalysed by open source culture.
The Open Climate Community Calls, launched by the well known Appropedia commmunity, created worldwide connections between multiple individuals from different domains – to understand the relationship between openness and climate change.
OS-Climate created another large-scale Linux Foundation collaboration between major companies to increase transparency and traceability in Environmental, Social, and Governance (ESG) ratings and support sustainable investment through an open source toolchain.
The Digital Public Goods Alliance is an initiative endorsed by the UN Secretary-General that facilitates the discovery and deployment of digital public goods. Similarly to the Open Sustainable Technology project, it provides an index of valuable open source projects related to the UN Sustainable Development Goals, accompanied by a billion-dollar investment commitment.
The Coalition for Digital Environment Sustainability (CODES), is an international multi-stakeholder alliance bringing together the scientific community, governmental institutions, NGOs, tech companies, and civil society to champion digital sustainability.
Fig. 4 - With the disclosure of the Landsat archives, the use of these satellite data increased noticeably Source: Landsat Archive Dashboard#
Definition of terms#
When it comes to complex (socio-technical-ecological) systems, many different perspectives can be taken depending on the context. This study examines the relationship between three intersecting dimensions – open source culture, technology, and environmental sustainability – from the perspective of ‘open sustainability’. This term must be clearly distinguished from other similar concepts relevant to this study.
Fig. 5 - Dimensions of open sustainability.#
Sustainability is a concept that is concerned with meeting the needs of the present without compromising the resources of future generations. Sustainable systems are those that “meet the current needs of many individuals involved in producing, deciding, and using a commons without compromising the ability of future generations also to meet their needs”.1 In this sense, sustainability is achieved “as long as the average rate of withdrawal does not exceed the average rate of replenishment”.1
Open sustainability refers to the use of open cultural and technical approaches towards sustainable outcomes. The concept examines how open source culture, technology and methods contributes to all three dimensions of sustainability – ecological, economic and social. It is concerned with how openness as a philosophy is instrumental to sustainability through the acceleration of transparent and collaborative innovation. Within this study, this concept is dealt with in detail, with a particular focus on environmental sustainability.
Previous definitions of open sustainability are broadly focused on open innovation within the context of sustainable development, often purely from an internal corporate perspective.2 We consider open source as an essential component of transparent and collaborative innovation, therefore, provide a more narrow definition that explicitly embodies open source culture and methods, without restricting its application to any one actor. Open source appropriate technology (OSAT), coined by the Open Design movement, is another related term. Appropriate technology – considered to be largely sustainable, small and appropriate – is within the scope of open sustainability. However, open sustainability is not prescriptive concerning the design specifications or implementation details of technologies. Instead, it defines methods and guiding principles from which similar technological attributes can emerge.
Digital sustainability is another related term, defined as the “process of applying social, economic, and environmental stewardship principles to digital products, services, and data delivered via the internet”.3 It is often included within digital transformation strategies. While digital sustainability is an aspect of open sustainability, open sustainability is predicated on openness and makes this explicit. Furthermore, digital sustainability is constrained to digital products, whereas open sustainability is not.
Sustainable technology is a broad term most commonly used to refer to clean energy sources and systems that minimise environmental impact. For example, so-called “green software”, which is concerned with lowering the energy consumption and carbon intensity of cloud computing, would be considered a sustainable technology. Whereas an energy-intensive climate model may not be sustainable in its operation, it can provide critical insight that informs sustainable decision-making with vast implications.
Open source sustainability refers to the ability of an open source project and its community to sustain themselves over a long period of time and to adapt the project to new circumstances and technologies. A strong community and governance structure is central to delivering bug fixes and new features.
Open source sustainability is widely known and analysed in detail within numerous books and publications.4 A whole ecosystem has emerged with the objective to commercialise the support, risk analysis and funding of open source software. Within the Linux Foundation, a new community emerged to determine the health state of an open source project: Community Health Analytics Open Source Software (CHAOSS). Another strong community that has formed in this space is Sustain. This work is important, as large, well-known projects typically receive more donations than small ones. However, small modular projects can be critical to the global digital infrastructure. Even though large companies use these projects, donations are small compared to the development resources that are required.5
Fig. 6 - A meme illustrating the dependency structures within software shows how even small projects can be of central importance for an entire software ecosystem. Source: xkcd#