Fostering a Digital Commons: Internet-Native Experiments For Sustainable Open-Source Software

The internet, and by extension much of the global economy, is built on “digital public goods,” often in the form of open-source software. This open-source software is a critical infrastructure for the devices and applications that are part of our everyday lives, yet remains consistently under-funded and under-resourced. Below we explore the importance of and obstacles to sustainability that open-source software faces today, and highlight various blockchain-based experiments that aim to improve the quantity and quality of such infrastructure.

Public Goods and Their Problems

The idea of public goods has evolved from simply defining what is “good for the public'' to a robust set of conceptual social and economic frameworks. Historically, philosophers like Jean-Jacques Rousseau focused on the role of states in provisioning public goods for society’s common benefit, while economists such as Paul Samuelson tried to further distill the term into notions of non-excludability and non-rivalry (i.e. a lighthouse is available to everyone but captured by no one) to help explain the unique properties that make these goods difficult to maintain. Using the widely adopted matrix that arose from these definitions, we might informally say that publics are a kind of expanded commons. 

In a world increasingly made up of networks, in addition to markets or states, technologists have begun to return their focus to the ways in which public goods produce positive externalities – that is, the ways in which they generate compounding value beyond just the communities they serve. A public park might have operating hours, and a subway system might charge a fee, but it’s clear that we can collectively benefit from the ways in which these goods help us support ourselves, connect with each other, and contribute to broader society. 

Across all of these definitions, the sustainability of public goods is always a common thread. Well-known challenges like the “free rider” problem and the “tragedy of the commons” are recurrent as individuals constantly seek to benefit from resources or services without contributing to them, resulting in an undue burden on those who do. In the case of the public park, picnic-goers might all enjoy the space, but only some will follow the rules and leave it clean. If few enough people follow this social contract, goods like our parks will inevitably fall into disrepair depriving everyone of the benefit. 

The State of Digital Public Goods

As public goods become increasingly multinational, the coordination and sustainability problems associated with them grow as well, often exponentially. On a global scale, these issues can appear more complex or even intractable as single states lose the interest in or capacity for ensuring continuous provisioning. In fact, in the worst cases, states can be incentivized to engage in exactly the same kind of free rider behavior as individuals at an institutional level. The consequence of this is that the most public goods suffer from coordination failures most acutely, despite producing even greater positive externalities as they reach larger populations.

Digital Public Goods as a particular class of global public goods take these challenges to their limits: not only are they fundamentally global in that they exist across all jurisdictions, they are “abstract” in the sense that they are not present or visible within any one of them. As researcher Nadia Asparouhova and others have noted, digital public goods use the unseen labor of internet citizens to produce some of the most useful but most invisible infrastructure we all rely on. In this sense, the internet can sometimes feel like its own special economic zone with its own needs, its own capacities, and its own arrangements with other jurisdictions (as others have so famously mentioned: code is law).

The social fabric that created digital public goods actually predates the internet (and the term itself) by quite some time, dating back to the early open-source software movement that rose to prominence in the 1980s and 1990s in response to the rise of proprietary software. Fundamentally, maintainers and contributors who grew up in the Unix and GNU (recursively – “GNU Not Unix”) era were deeply influenced by Richard Stallman’s introduction of the GNU General Public License (GPL) and strongly objected to the increasing commercialization of code, reduction in collaboration, and general restriction of user freedoms. 

As tooling like Git in 2005 and later more centralized platforms such as GitHub grew in popularity, open-source as a philosophy began to garner significant interest from the broader developer community and eventually from the traditional institutions who had previously rejected it as inefficient and uncompetitive. Slowly, these groups realized that by sharing code and building collaboratively (much like academic institutions did in the early days of Silicon Valley) they could arrive at more innovative solutions to development problems and produce higher quality software. Today, open-source values such as peer review, open code contribution, and community feedback have become enshrined as obvious drivers of innovation and quality, while licensing initiatives (e.g. GPL, MIT) have continued to maintain and enforce these ideals. 

Given its success, international bodies such as the UN have placed priority on finding new ways to continue support for these types of initiatives under the banner of digital public goods. The recent Roadmap for Digital Cooperation states: “If the benefits of increased internet connectivity are to be realized, it is important that all actors, including Member States, the United Nations system, the private sector and other stakeholders, promote open-source software, open data, open artificial intelligence models, open standards and open content that adhere to privacy and other applicable international and domestic laws, standards and best practices and do no harm." However, the level of support these initiatives are currently providing still pales in comparison to the financial value (sometimes upwards of $300B) states gain from the work of these unique special economic zones.

As a result, the question that many philanthropists and technologists have begun to ask is: what other models can we explore to help sustain this unique class of public goods given the value they continue to create?

An Introduction to Decentralized Autonomous Organizations

Many technologists who have looked into the problems of sustainability and equality in open-source have begun to recognize that it might be possible to fix them using the very properties that made the movement so successful in the first place. Just as the reduced cost of transferring information through the internet helped enable new forms of code collaboration (i.e. version control), the guarantees of consensus-based distributed systems (i.e. blockchains) are creating a new wave of experiments around transferring value through shared verifiable commitments. In this sense, like the internet, blockchains are inherently generalizable and trustful systems for all kinds of collective coordination.

Perhaps the most interesting applications of blockchains thus far have been in the form of Decentralized Autonomous Organizations (DAOs). These distributed internet-native organizations use verifiable commitments to maintain global records of contributor actions and governance ownership as they work together towards shared goals often entirely online. While some of the solutions offered by these new systems are not necessarily novel, many of these organizations have sped through existing institutional design spaces to both build on implementations of old ideas (e.g. platform cooperatives) and plant the seeds of new ones (e.g. globally organized currencies).

One particularly relevant area of exploration in the DAO landscape today is around how to effectively reward contributors in ways that encourage more equitable ownership. To this end, some DAOs have even opted to issue their own community currencies in hopes of reducing freeriding while balancing opportunities for voice and exit. In these kinds of organizations, those who contribute code, provide support, or add value are able to take part in the governance of these organizations and in some cases benefit from the adoption and use of the projects they are part of, fostering a more sustainable ecosystem.

Internet-Native Experiments to Improve Digital Public Goods

Especially over the course of the pandemic, many open-source contributors began creating their own internet-native organizations of the kinds we have described. Below, we highlight just a few key experiments that are leveraging blockchains and other kinds of distributed systems for value creation and distribution.

Gitcoin

Gitcoin is a DAO focused on building and funding digital public goods. By deploying an instance of their open-source protocol, any community can leverage the trustful properties of blockchains to create participatory, open grants programs for its members. To date, Gitcoin has tested its tooling in over one hundred distinct grants pools that have collectively distributed over $50M in partnership with organizations like the Ethereum Foundation and UNICEF. A few notable projects that have been funded include Electronic Frontier Foundation, Internet Archive, MetaGov, Uniswap, and WalletConnect.

A primary contribution that Gitcoin has made to the open-source funding landscape is popularizing quadratic voting as a methodology. Using an extension of the concept known as quadratic funding, they have helped show that measuring the intensity of voter preferences (as opposed to methods like ranked-choice) can lead to a broader diversity of grantee allocation outcomes. In addition to helping find outliers, quadratic funding allows larger philanthropic and institutional backers to directly contribute to what their communities care about by donating to matching pools focused on a specific cause area (e.g. ethereum infrastructure), simplifying the process of participatory grantmaking.

Since quadratic funding can be sybil-attacked, Gitcoin’s experiments have become fertile ground for testing different identity and bribery protection models that are widely applicable to distributed applications at large (e.g. IoT). As more organizations use their protocol to deploy open grants programs, anonymized research is published around results to improve these models for other rounds and other projects in the future.

Through its own community-led governance, Gitcoin is able to apply the ideas it promotes to itself and reinforce the equitability and sustainability of its mission through self-infrastructuring. This reflexivity makes it an interesting case to study as its internal tools continue to evolve.

Radicle

Radicle is a decentralized code collaboration network that integrates distributed identity features and Git's version control system. Each peer maintains a branched version of the project, and identities are created with a key pair from the command-line interface (CLI) and can be associated with a “.eth” domain. 

Most of the open-source code today resides on centrally hosted platforms, which are predominantly owned by corporations such as GitHub and GitLab. These platforms, due to their corporate ownership, are subject to corporate laws and regulations that can impose certain restrictions and guidelines. They are also potentially vulnerable to censorship, due to the need to comply with laws and regulations, the company's own policies, or other external pressures. This centralized structure can at times limit the free and open distribution of open-source code, which is one of the cornerstones of the open-source philosophy. 

The key distinction between Radicle and traditional Git workflows lies in the underlying network that powers them. Radicle's network relies on Radicle Link, a peer-to-peer replication protocol built on Git. Unlike the traditional 'cathedral' model where there is a single canonical 'master' branch that contributors merge into, Radicle Link fosters a 'bazaar-style' collaboration model. In the conventional model, where there's a reliance on a single master branch, disruptions such as server downtime or a single point of failure can halt progress or even risk data loss. With each peer maintaining a branched version in the Radicle model, the project becomes inherently more resilient as the codebase is decentralized and redundantly stored. Each peer can choose which changes they wish to keep, fostering an environment of truly collaborative and flexible development. 

Radicle projects can also have multiple approved signers, creating a multi-signature model where more than one delegate needs to approve changes to the project. By requiring more than one approved signer to validate changes to a project, it reduces the likelihood of unilaterally imposed changes and encourages a broader involvement in decision-making. This can enhance the cooperative nature of open-source projects and give a voice to contributors who may otherwise be sidelined in more centralized models. While seed nodes are currently necessary for publishing changes, the platform is working towards a fully built-out peer-to-peer replication system. This will provide additional redundancy, ensuring that the codebase remains accessible even if a seed node goes offline.

Furthermore, by leveraging blockchain-based tokens and smart contracts, Radicle has the potential to enable innovative, decentralized funding models for open source projects. It could allow contributors to receive fair compensation for their work, addressing a long-standing challenge in the open source community.

Hypercerts

Hypercerts are a newly proposed interoperable data layer for impact funding in which each individual certificate describes an impact claim about a specific scope of work, for example an open-source software library. Their goal is to test on-chain solutions to long standing impact measurement problems (i.e. how contributors measure the value they have created) using a market-based approach in which tradable certificates are purchased and proposed outcomes are tracked, measured, and achieved.

While Hypercerts do not seek to fully “solve” the difficult task of impact evaluation, putting impact evaluations into a formally specified data layer allows individual ecosystems to construct networks of impact claims and more effectively form impact markets. The Hypercert specification allows for these networks to be open and participatory, or more often for them to be expert-driven (i.e. in which domain-specific evaluators curate the certificates on the market). 

If successful, Hypercerts would potentially open up forms of impact investing to a broader class of participants, including new evaluators, who could be rewarded for their role in facilitating the discovery of new projects that are not easily visible to the community. In turn, the unseen labor of open-source core contributors could be made more visible.

SourceCred

SourceCred was an open-source project that aimed to “make the labor of individuals online more visible and rewardable as they work together in a project or community”. Contributors to a given instance of the system would be assigned a non-transferable token called "Cred" using community-defined weights that determined the value of a set of actions, creating a network graph of flows of value. Over time, this would in theory allow the community to prioritize what it values the most. New tokens for a given community that are transferable could be distributed to those with Cred, redeemable for stablecoins like DAI or USDC so that these value flows could actually pay the bills.

This approach tried to solve the challenges of reputation and compensation that often exist within open-source projects. While the SourceCred community has undergone several governance changes, and was the subject of an ethnographic case study on blockchain governance, an updated version of their tooling is still in use by other decentralized organizations.

Recently, projects like Gov4Git have emerged that could improve on this approach and help it to reach a wider audience by solving some of the identity, reputation, and contribution prioritization issues that hindered SourceCred in its initial attempts.

Conclusion

Digital public goods, including open-source software, play a crucial role in connecting us to our communities, supporting economic activity, and improving the way we coordinate with one another towards shared agreements. Given their global nature, these goods produce significant value regardless of their geographic location or jurisdiction. However, for the same reasons that these goods are so generative, they face unique challenges around sustainability. 

DAOs and other internet-native experiments demonstrate promise in addressing some of the underlying issues open-source software faces today. By establishing mechanisms that better reward contributions, improve opportunities for equitable ownership, and more generally align incentives between labor and capital they can encourage continued interest not only in building but also maintaining these key technologies.

To sustain the immense value that open-source software brings to society, it is crucial for philanthropists, technologists, and international bodies to explore and support these different models and methods to create better outcomes, especially for labor, in these unique environments. Especially in the face of a rising number of global public goods problems, one of the best investments we can make is in our shared infrastructure.

Sarah Hubbard is currently a Senior Fellow at the Harvard Kennedy School focused on emerging technology and policy. She is a technologist and product leader who has led various teams building machine learning tools and intelligent devices at Microsoft and Apple.

Scott Moore is a Co-Founder of Gitcoin, an internet-native community for building and funding digital public goods. He's also an active steward of related projects including ENS, Gnosis Safe, and Optimism. As a researcher, he explores topics related to collective intelligence, equitable governance, and impact funding alongside MetaGov.

Helena Rong is a PhD Candidate in Urban Planning at Columbia University and Founder at CIVIS Design and Advisory, whose work focuses on the intersection of design, technology, and urban research.

Anand is an entrepreneur-turned-public interest technologist with diverse experiences in responsible commercialization of technology and digital development. His special interests include Open Digital Ecosystems, AI Governance, and Web3 for impact.