Priorities to Make the Fediverse SustainableAlek Tarkowski / Dec 9, 2022
Alek Tarkowski is the Director of Strategy at Open Future.
When Elon Musk purchased Twitter in October, the #TwitterMigration began. Users started setting up accounts on Mastodon, a social network that is close in functionality, and part of the decentralized “Fediverse.” User count started growing quickly on October 28th, and by November 20th the Mastodon network grew from 5.7 to 7.9 million users (with a continuing weekly growth rate of around 400 thousand).
Mastodon still is equal to only a few percent of Twitter's user base, so the migration is not significant in absolute numbers. These events are unique for a different reason: they are a case of collective user choice over the network they use, benefitting from the availability of a decentralized solution and fuelled by an ethical, and not just consumer choice.
As such, this is worth paying attention by those interested in building sustainable, large-scale digital public spaces. An emerging network of several million users, even if dwarfed by corporate networks, represents a huge opportunity to experiment and test ideas about the design and deployment of online public spaces.
Is the Fediverse the digital public space that we need? Is it an alternative to corporate platforms, and can it sustain networks of users hundred times bigger than its current size? On one hand, Mastodon is already a robust, decentralized network that functions in contrast to the dominant platforms: there’s no centralized control, no adtech, and no exploitation of user data. On the other hand, it faces two major limitations.
One has to do with its sustainability, as the commons-based peer production model on which it is built has clear limits to growth. And the second, and perhaps most crucial, has to do with a lack of participatory governance. The Fediverse is ruled today through code, by a narrow group of its maintainers: developers of code and administrators of servers. Just like in the case of the production model, this approach – well known to those familiar with open source projects – is paradoxically both the source of Fediverse’s resilience, and a significant limit on further growth.
Newbie users, and a newbie network
As new users set up their accounts and start using Mastodon, they realize that some things work differently than on Twitter, and sometimes not as smoothly. There are thousands of servers to choose from. Search works in a weird way. DMs are complicated, and there are concerns about their security. As a result, multiple guides are being written for newbies (such as this excellent one, from Maya.Land).
These guides for newbies usually frame the challenges that the Fediverse faces today as related to the individual behavior of users. There’s advice on how new users can educate themselves about the functionalities of Mastodon, suggestions for users to better distribute themselves across servers, requests to crowdfund the upkeep of servers, and admonitions to follow established rules and norms. Handling the sudden growth of the network is framed as a matter of individual choices.
Yet Mastodon faces not just over two million (and counting) of newbie users. It is also today a “newbie network”. The influx of newcomers means that we are facing a new Mastodon (and a new Fediverse). The challenge is not just one of scaling up the network, but a cultural one. The network is growing beyond the original user base of hackers and geeks that have established a stable presence in the Fediverse in the last five years or so.
In this context, much of the “newbie” language is almost dangerous, as it frames the challenge in terms of individuals needing to learn technical skills, and also to comply with an existing culture. Instead, longtime Fediverse users and maintainers need to think more about what they can learn from the “newbie network”. The network needs to learn about itself as it grows and transforms, and to further develop its infrastructure, its institutions, and its political economy. And the challenges that it faces lie not at the level of individual users, but at three key layers of the Fediverse stack: the servers, the services and the protocol.
Why does the Fediverse need to grow?
Why does the #Fediverse need to grow? I was asked this question after sharing some initial thoughts on my Mastodon account:
@tarkowski why do we need to instrument the network, if it is working for users? There are no shareholders demanding growth, here.
This is a fair point. Mastodon is a sustainable, healthy network that reached – before the migration begun – an equilibrium of around 5 million overall, with half a million active users. So why does it need to grow further? Because millions more people need access to healthy, just, sustainable, user-friendly communication tools. Hans Gerwitz described it as seeing the network’s growth as “souls saved,” instead of “eyeballs captured.”
French sociologist Alain Touraine, one of the great theorists of social movements, defined movements as “organized behavior of a class actor struggling against his class adversary for the social control of historicity in a concrete community.” Historicity is the term Touraine uses to describe the ability to manage major orientations of collective life: the directions that our societies take and changes that shape them. This might sound grandiose, but it does put current events in the right perspective. Ultimately, what is at stake is our capacity, as societies, to control and manage planetary-scale communication spaces that are, in the broadest sense, democratic.
If there are to be meaningful alternatives to corporate walled gardens, and if we are to see users move not from one such garden to another, but also to spaces that are civic and public in nature, the alternatives need to be more than just a niche.
Yet that’s what the Mastodon network is today, even taking into account the current influx of users. In this regard it is similar to many other progressive tech ideas – like data cooperatives, for example. Data cooperatives have spawned abundant theoretical literature and a healthy dose of activism and experimentation, but only on the very margin of the commercial, private solutions with which these initiatives compete.
The Twitter Migration happened because of an external factor. Without Musk’s takeover as a triggering event, one million of users would likely not have created Mastodon accounts. And it’s easy to find examples of earlier migrations: abandoned accounts of well wishing users, giving up after several toots. This points again to the limiting factors that Mastodon faces: the combined issues of network sustainability and governance.
The frugality, and the precariousness, of open source
“I do software development, devops, accounting, customer support, project management, product design, public relations, and moderation for 36K per year…”, Mastodon creator Eugen Rochko wrote on October 31st.
Eugen, known in the Fediverse by his handle, @Gargron, is the founder, CEO and lead developer of Mastodon GMbh, the German nonprofit company that develops and owns the code of Mastodon, the dominant service in the Fediverse network. Eugen’s story is an amazing one of personal vision and commitment. He has almost single-handedly built Mastodon for over five years. In Polish we call such people “człowiek instytucja” - “a person-institution”, to indicate that their individual value to a community or society at large rivals that of whole organizations.
Available data on the finances and sustainability of Mastodon shows the amazing frugality of its creator. On the other hand, it hints at the precarity of the setup. Eugen admits himself that his wage is below the average wage in Germany, where he lives. While one can argue that money is not the only, or necessary, driver, crucial socio-technical projects should be well resourced. Financial data made available by Mastodon suggests it operates on a yearly budget of around 100K Euros. This includes two public grants, one for 9K Euro from the German Prototype Fund and another for 45K Euro from the European NGI Zero program.
It’s hard not to put this number in perspective by comparing it with typical levels of funding for commercial startups, and to recognize the role of European public funds. Thinking about the relevance of the Fediverse today, and of the repeated calls for public interest alternatives, it is hard to understand why the Fediverse is not properly resourced, either by public funding or by charities.
This problem is not new. For years now the open source ecosystem “has had a funding problem”, as James Turner wrote on the Stack Overflow blog. There are plenty of examples, such as that of cURL, an important software library that has been developed by a single developer, for almost three decades, almost for free.
This is a case of what Paul Keller and I have called the Paradox of Open: the existence of power imbalances that leads to, at best, ambivalent outcomes of openness. It is a paradox that the success of open code software, both in terms of the reach of the technology and in economic terms, has happened through underfunded or entirely volunteer work of individual coders. This shows the limits of the open source development model that will affect the future growth of the Fediverse as well.
In his recent piece about the Fediverse and the “Post-Musk Net”, CUNY Newmark Graduate School of Journalism professor Jeff Jarvis expresses hope that in the “distributed future” individuals, institutions and entrepreneurs will experiment, innovate and offer other, new services and functionalities. Yet there are surprisingly few innovations in the Fediverse, a space that in principle should be generative. For example, it is surprising that there have been no prominent attempts to build decentralized, responsible algorithmic tools that would manage content for and by users.
One can, of course, argue that the network simply does not need to follow a growth imperative. But additional functionalities, and a robust process of innovation and development, are needed to keep this ecosystem healthy and sustainable in the long run. Ensuring this is a matter of supporting additional capacity to develop and maintain the network.
Governance beyond benevolent dictatorship and democracy of servers
The production model of Mastodon also has consequences for its governance. GitHub statistics show that @Gargron has single-handedly produced and committed at least 80% of Mastodon’s code. But more importantly, as the lead maintainer of the code, Rochko makes all the decisions. In line with a typical open source model, where commitment determines stakes in the governance process, the shape of the service is decided by one person, a self-described “benevolent dictator.”
One can say that this dictatorship has served well a growing network of half a million happy users. But it’s hard not to think that this model is far from the ideal expected of democratically managed and governed infrastructures. One telling example is the lack of a functionality similar to quote retweets, a feature that’s popular on Twitter but missing on Mastodon. The reason for this is that Rochko has a strong belief that quote retweets mainly serve to encourage negative forms of engagement, such as hate speech; he has therefore not implemented them. A real community-driven process, or at least a consultation on this key issue, would be a good starting point for expanding the governance of Mastodon.
Of course, one can make the argument that both the Mastodon code base and the underlying ActivityPub protocol are open, and thus others are free to experiment. Everyone is free to copy the code, and to launch alternative apps in the ActivityPub-powered Fediverse, competing in this federated network. But it is an argument that fails account for network effects that are at play even in relatively small networks of several million users. And, it does not resolve the fundamental inequality inherent in open source systems between those who code, and those who cannot or do not.
The same factors are at play at the server layer. Since both Mastodon and the larger Fediverse are decentralized, they are open ecosystems, in which anyone can set up and maintain a node: a server running a Mastodon instance, federated through the ActivityPub protocol with the rest of the Fediverse. Over the last two weeks the number of instances grew proportionally more quickly than the number of users, showing that the model can sustainably scale. At the same time, many servers had to limit or stop accepting users, as either the technical infrastructure or their finances broke.
But servers are more than just ways to distribute load. Servers are also nodes that control information flows, through content blocking decisions. Unlike on the world wide web, federation does not mean full interconnection. The Fediverse is, by design, fractured by server-level decisions that block and cancel access to other parts of the network. Users can also make such decisions at individual level, but at the scale of the network, it’s the servers that count.
Servers are typically governed by individual admins, who maintain and often fund their own operations, sometimes with the support of users. Although there is no server governance census, in my own research I found very few examples of more participatory approaches to server governance (social.coop keeps a list of collectively owned instances). One might argue the status quo is working: there has been no significant popular revolt of users against admins. Still, governance of instances – and specifically content moderation decisions – are one more level at which more participatory governance could be introduced.
The crucial divide, just as with the Mastodon code, is between the programming haves and have-nots, the coders and non-coders. The openness of the ecosystem means that it is in principle a lot more democratic, as it creates meaningful possibilities to shape it by contributing code. But this ability is not available to the majority of users, leading to a sort of caste society, built on top of an open source infrastructure. There is no realistic scenario in which all users learn to code – therefore participatory governance approaches, which take control of the code away from the hands of the coders, and into collective decision-making processes, is the only way forward.
These dynamics also happen at the level of the Fediverse as a whole. The core protocol, ActivityPub, is a W3C standard, and therefore based on a strong approach to participatory governance – although one limited to a very small group of people with the interest and capacity to contribute on a voluntary basis to a complex process of technical design. And once the standard has been adopted, the governance processes at this layer have slowed down, and then stopped.
Better participatory governance is therefore the big challenge ahead of Mastodon, and the Fediverse space more generally. Web3 is usually detested by open source developers, and this negative vibe is also strong on Mastodon. Yet it is in the progressive margins of the Web3 space that debates on governance models take place. Successful governance experiments, conducted for example around the Gitcoin protocol, or the theoretical work of the Metagov project, should be examined and applied to the Fediverse. And of course it is not just the Dweb space that can serve as an example. There are many participatory frameworks that could be experimented with, and deployed, on the solid root of the open and communitarian ethos of the Fediverse: Decidim decision-making, or advanced participatory models like citizen panels.
The key transformation that needs to occur is from a logic of open source development to a more complex logic of commons-based governance. Jan Zygmuntowski and I recently published a primer for data commons initiatives. Our blueprint for such institutions includes three core parts: stewarding access, generating public value and collective governance. The Fediverse, as it exists today, has solved the first challenge, and is relatively good at generating public value (though more could be done to ensure not just sustainability, but greater productive capacity). The most work is needed to ensure collective governance in this space.
The mission of building digital public spaces
The challenge with governance is that it requires collective action - unlike coding, which can successfully be done, on behalf of millions, by a single coder. Governance also requires a different approach than the Twitter migration process - which happens through a sum of decisions and small steps taken by individual users. In sum, it requires a conversation to take place. This conversation does not happen because of coordination problems that such communication entails, coupled with a belief in governance through the peer production of code, on which open source development is based.
For this reason, I am uncertain whether such a shift towards participatory governance is possible. A useful analogy is that of Wikipedia and other Wikimedia projects, which are undergoing a significant “phase shift”, from the culture defined by the community of early contributors, to a broader and more inclusive culture– one centered not just on encyclopedic prowess, but also institutional organizing. This example suggests that such a shift is possible, but hard. It requires both significant resources, which have been invested in the case of Wikimedia, but also strong leadership that is in dialogue with the community and can negotiate together the changes (this has happened to a lesser extent).
The Fediverse represents a unique opportunity. It represents a chance not just for this particular network, but for all the visions of a better, public interest-driven, “for good” internet. The “newbie network” that emerged in recent weeks is a mix of a sustainable network with a strong open source ethos, combined with the influx of new ideas, critiques, capacities and resources brought by a wave of users with a strong interest in developing better alternatives. The question becomes, are we able to go together on this mission?
To start on this path, I recommend three priorities:
- Launch a participatory project to define a shared mission for building the digital public space on the basis of the Fediverse. There are multiple methodologies to do so, such as Decidim, deployed in urban settings; or Pol.is, participatory experiments conducted by the Taiwanese government. The Conference on the Future of Europe, a EU wide deliberative process, provided surprisingly good results as well.
- Secure greater involvement of public institutions. Dan Hon proposed for organizations to set up their own Mastodon instances and serve as verified, public interest driven, trusted nodes in the network. Public institutions should also bring in resources needed to develop the network - for example invest in public benefit algorithmic solutions for the Fediverse, or a broader range of services (Peer Tube is a good example of a publicly funded Fediverse service).
- Build a stronger social and institutional layer. There are many experts in community- and network-building who should be interested in working on such a project. There are also great prior examples of large-scale, commons-based peer production (the BigScience and BigCode projects are the latest examples on my radar).
The growth dynamic from last month has slowed down significantly. And we can expect the emotional rush of joining new communities, and trying out new tools, to cool down soon as well. This is the time to start thinking about the long-term sustainability and governance of the new, bigger Mastodon. Not because there is a growth imperative, but because the world needs proof that social networks can be maintained and governed in a decentralized and democratic way.