What we’re growing at Oxford Boaters is simple

From the towpath at dawn to moonlit moorings at dusk, Oxford’s boating community is not a curiosity, it’s part of the city’s living fabric. Generations of people have chosen to make their homes on the water, creating a culture rooted in community, care, and independence. This is a quieter Oxford, rarely captured in guidebooks but felt by anyone who walks the river or canal: a human-scale world of shared tea, passing conversations, and everyday presence. In a city shaped by wealth and exclusion, the river remains one of the few places where real social diversity still exists, where people live side by side not because of status, but because they’ve chosen a different way of life.

This matters because the boating community doesn’t just live on the river, it sustains it. Their daily presence keeps the towpath safer, more welcoming, and alive. They act as informal stewards, noticing changes in the water, caring for the banks, and maintaining a relationship with the environment that no institution can replicate. Remove this community, and you don’t just lose homes, you lose Oxford’s character, its openness, its lived connection to the river. The waterway becomes quieter, more managed, less human. What looks, from a distance, like a marginal issue is in central, a question of whether #Oxford remains a living city, or becomes more controlled, polished and diminished.

How can we protect that space?

We want to protect a shared space, keep the river livable, and organize ourselves to have a voice. None of that is technically complicated, the difficulty isn’t the goal, it’s the people, and the tools we use to try and work together. No matter what process we choose, it always comes back to the people who make it work, or the people make it stall. Some processes recognise this and work with human reality – trust, conflict, misunderstanding, ego, care. Other processes ignore it, and end up being used (consciously or not) as blocking. Let’s look at a few grounded examples of this.

We’ve already seen this clearly, in a face-to-face meeting (the “whispering fire”), trust rises fast. People read each other, soften, find common ground. You can go from 50% to 70–90% trust in a couple of hours. Move the same conversation to online chat, and trust collapses. Tone gets lost, small disagreements escalate, and people start pulling things apart. You drop to 20–30% trust, sometimes lower. Completely different outcomes, it’s not a failure of individuals, it’s a mismatch between tools and human communication.

The website vs the chat is another clear split. The website (or any structured space) holds higher trust, but lower participation. The chat holds high participation, but low trust – knowledge isn’t captured properly, decisions aren’t visible and new people can’t easily get up to speed. Result: constant rehashing, frustration, and burnout leading to momentum loss.

There’s a temptation to design the perfect structure with formal agendas, strict procedures and detailed governance. On paper, this looks like progress, in practice, it becomes a brake. As people use “hard” process to delay decisions (“we need another meeting”), avoid responsibility (“that’s not my role”) and assert control (“this isn’t the proper channel”). Instead of enabling action, the process becomes a gatekeeper, leading to the same basic issues resurface again and again. Not because people are stupid, because the basic social fabric isn’t being maintained.

Processes that work can see this cycle and design around it so as not to keep restarting from zero. So what actually works? The path isn’t finding a “perfect” process, it’s choosing #4opens processes that fit people as they are. That usually means prioritising face-to-face (or close equivalents) for trust building. Keeping structures simple (#KISS) so they don’t become tools of control. Capturing shared knowledge clearly (FAQ, summaries, decisions). Accepting mess as normal, but making sure it composts rather than festers. Balancing fluffy and spiky – you need both to move forward

And most importantly recognizing that process is never neutral, every structure we put in place will either help people collaborate or give them ways to block each other. Often both at the same time. So yes, what #Oxfordboaters is trying to do is on the surface easy, but what makes it hard is human complexity, mismatched tools and blinded pushing processes that don’t align with either. When we get those bits even slightly more right, everything else becomes possible.

If we don’t, even the simplest goals turn into a grind, that’s the real work.

Closed systems protect individuals, but they rarely build movements

Trust and fear – the spiky/fluff debate – OxfordBoaters

Closed systems protect individuals, but they rarely build movements

People fight against or/and ignore the #KISS (Keep It Simple Stupid) approach in tech because simplicity exposes power. Complexity, jargon, and process give cover – they make control look like competence. When paths are simple and transparent, everyone can see who’s blocking, who’s hoarding, who’s acting in bad faith. Many “experts” and institutions are emotionally and professionally invested in keeping things complicated; simplicity threatens their authority, their funding, and their identity.

All the #OMN projects are not directly about social change – they’re about making social change possible. That distinction matters as people don’t step into change unless they first believe change can happen. If the world feels fixed, locked, inevitable, then nothing moves. Our role is simpler, and maybe more important, to open that door a crack, to show that different paths exist.

Think of #OMN as a helping hand, not dragging people forward, not telling them what to do – just making it easier for them to take that first step when they’re ready. But to do this, we need to think more clearly – and more fundamentally – about technology itself. As most of the current “open paths” are cosplay at best, we need a network that links them as flows for there use to be unlocked from the current limits of #stupidindividualism shaping them – to become a native part of the expanding #openweb reboot.

I’ve been working on this for over 20 years, and one thing keeps proving true: we need roughly 90% open and 10% closed, the balance matters. As the current push from the #encryptionists flips this – aiming for 90% closed and 10% open. That isn’t a solution, it’s a retreat. It breaks the social fabric that makes collective tools usable and meaningful. It fragments, isolates, and ultimately shrinks the space where shared culture can exist.

Yes, privacy matters, yes, some things should be closed, that’s the 10%. But the commons – the space where we meet, talk, organise, and build trust – has to be open. Without that, there is no network, just silos. Take a simple example: you’re reading this via #activitypub. That’s a system built on being mostly open, with just enough closure to function safely. And it works, people are here, conversations happen, networks grow.

Compare that to more closed, encryption-heavy systems like old school Diaspora. Technically interesting, sure, but socially? Empty, few people, little flow, no impact. That’s the core point: this isn’t just about functions or features, it’s about culture.

Open federated, networked systems create the possibility of shared culture, and from that, the possibility of social change. Closed systems protect individuals, but they rarely build movements. We need both – but we need to get the balance right. Right now, too many people are getting it the wrong way round.

This Isn’t New: Decentralisation Was the Point All Along

Decentralised servers – what we now call the #Fediverse – are often talked about as if they’re some new, radical innovation. They’re not, they’re a return to the original design of the network. The early internet wasn’t built to be controlled, it was built to survive. The core idea was simple: if parts of the network were destroyed – even something as extreme as a nuclear strike – the rest would keep functioning. No centre, single point of failure or “off switch.”

That’s what decentralisation actually means. And this thinking didn’t even start with the #openweb. Systems like Usenet already embodied this approach: distributed, federated, run by many, owned by none. Messy? Yes. But resilient, open, and hard to capture.

What we’ve been living through for the last 20+ years – the rise of the #dotcons – is the opposite of this. Centralised platforms with single points of control. Easy to use for control and monetise, easy to manipulate, easy to shut down. We didn’t lose the #openweb by accident, we blindly traded it away for this convenience.

What we’re seeing now with the #Fediverse, #ActivityPub, and related projects isn’t innovation in the common sense. It’s a reboot, a return to the path we were on before we derailed it. The difference is that now we’re trying to rebuild this in a world that has spent decades normalising centralisation and control. 40 years of death cult worship has changed people, institutions, social groups and our very internal selves. That’s where the friction comes from, people arrive expecting #dotcons platforms, what they find is networks. People expect control, what they get is responsibility. People expect “free” what they face is shared cost and care.

So, it was never about the tech, the mistake we keep making is ONLY thinking this is a technical shift, it’s not, it’s cultural. You can spin up a decentralised server in minutes, that’s not the hard part, the hard part is everything around it:

Who runs it

Who pays for it

How decisions are made

How conflict is handled

How trust is built and maintained

This is the work the #dotcons hide from us, they wrap control as “free services” paid for with surveillance, extraction, and control. Now that we’re back on the #openweb path, that work becomes visible again, and yes – it’s harder.

Why this matters (Again). Resilience isn’t an abstract idea anymore as we’re living through cascading crises: political instability, #climatechaos, infrastructure fragility. A centralised network fails catastrophically were a decentralised network degrades – but keeps going. That’s the difference between a system you depend on and a system you can trust.

We don’t need to overcomplicate this – Keep It Simple (#KISS)

The #openweb is decentralised

The #dotcons are centralised

One builds commons, the other extracts value, everything else is detail. And yes nobody thinks the Fediverse is not messy, uneven, (yet) match the polish of corporate platforms. That’s fine, mess is where growth happens – if we compost it properly.

The #OMN view, we’re not trying to invent something new. We’re trying to make what already works usable at scale for media, trust, and collective action. The infrastructure is there, the protocols exist, the history is long. What’s missing is the shared layer – the commons – where information flows in ways people can actually rely on, that’s what we’re building.

If decentralisation feels radical, it’s only because we’ve spent so long inside systems that forgot #OMN #openweb #KISS

Trust and fear – the spiky/fluff debate – OxfordBoaters

People fight against or/and ignore the #KISS (Keep It Simple Stupid) approach to social movements because simplicity can expose power. Complexity gives cover, making control look like competence. When paths are simple and transparent, everyone can see who’s blocking, who’s hoarding, who’s acting in bad faith. Many “experts” and institutions are emotionally and professionally invested in keeping things complicated as simplicity threatens their authority and their identity.

As second step in change and challenge we need to face some uncomfortable roots: fear of death, fear of the “other.” Fear, as the man said, is the mind killer. Too few people trying to change the world bother to look at the psychological ground they’re standing on. Our social mess isn’t random, it grows from somewhere. If we don’t deal with the roots, we just keep trimming the leaves and wondering why the weeds grow back.

At the centre of this is trust, it is what makes social change possible. It’s the messy, ongoing tussle between horizontal and vertical forces, between collective process and imposed control. Take Oxford boaters as a lived example, this is observational, based on 40 years of doing this kind of work.

Face-to-face meetings? They start at maybe 50% trust and can rise to 70% without too much trouble. That’s because the people who show up are self-selecting – a mix of fluffy and spiky, but open enough to engage. You get debate, but also movement.

Then we go back online were trust drops fast – down to 20–30%, sometimes even into negative territory. That becomes the dominant tone as things splinter. Work that was built collaboratively offline gets pulled apart, the focus doesn’t hold.

On the open collective website, trust is actually high – but usage is low, again, self-selecting. So rollout stalls, no shared space, no shared understanding, no momentum.

So we call another face-to-face, the “whisper fire.” Trust shoots up – 90%, easy. People align, decisions get made, it feels like progress. But then everyone goes home, the next day? Momentum evaporates, as few people feel responsible for carrying things forward. And very quickly, we slide back into the low-trust dynamics of online chat.

Meanwhile, the group has grown, more people from offline outreach and leafleting. But the website is still stalled, so there’s nowhere to hold shared knowledge. We keep re-arguing basics, we don’t even have a solid FAQ. Trust drops again.

Next meeting: bigger pool, smaller turnout. Trust starts at near zero, the first half is rough – people filtering out, clashing, posturing. Slowly, some shared ground emerges and trust crawls up to maybe 50% but still, nothing concrete gets resolved. A few working groups are seeded. Environment sort of functions. Media splinters from the start, but manages a press release, but without any clear #4opens process. Beyond that, the next steps remain unanswered.

Back online? Trust gets ripped apart again. The whisper fire is supposed to be every week, but we have lost focus, but we try agen – half-heartedly at first. Poor turnout, then people drift in after a couple of hours. Trust rebuilds to around 70%. Real decisions get made, consensus emerges, it starts to feel like an actual working affinity group.

But the chat? Still toxic, low trust, constant tearing-down, fixation on side issues, and a push toward rushed bureaucratic structures that crumble under their own weight. The only thing that actually works is the consensus built in the whisper fire – because that’s what people really agree on, underneath the passive-aggressive noise. But after each cycle, trust in the online chat drops again, down below 30% on the surface. Maybe 40% underneath, if you’re generous.

And now we hit the next stage: formal process, bureaucracy, decisions that actually matter, where it gets real. And because online trust is still so low, everything becomes harder than it needs to be, friction everywhere, misunderstanding as default.

But – and this is the hopeful bit – offline trust is slowly growing. So maybe, just maybe, we can carry that through, if we don’t let the chat tear it apart first.

Individual fears

So how do we mediate this fear?

Process is what gives legitimacy

#OxfordBoaters

We keep making mess, then wondering why everything smells

A big part of this is the language we use, when we unthinkingly spread #mainstreaming terms, we push the worldview that comes with them, and that worldview is usually rooted in fear, control, and market logic.

Take “digital sovereignty.” it sounds solid, sensible, progressive. But, it’s a made-up term trying to frame the internet in nation-state and market terms – ownership, borders, competition. It’s a liberal answer to a fear-based economy: “how do we control this thing so it doesn’t threaten us?” That framing is the problem, because the #openweb was never built on control. It was built on trust, shared standards, and open process – the #4opens:

open data
open source
open standards
open process

That’s the native soil, when we blindly shift to language like “sovereignty,” we drag in assumptions that don’t belong. We start thinking in terms of ownership instead of participation, control instead of collaboration. And that creates mess – conceptual, political, and technical Then we spend years trying to “fix” that mess, composting it.

But it’s much better if we don’t make that mess in the first place #KISS, by staying grounded in #openweb values. We don’t need to retrofit control structures onto something that was designed to work without them. We don’t need layers of governance theatre to simulate trust, we can build trust directly through open processes.

This is why clarity, clear language matters, if we keep pushing borrowed language from the #deathcult, we’ll keep rebuilding its logic, no matter how good our intentions are. So yes, we need to talk more clearly about stopping importing broken concepts, stop framing open systems in closed terms and stop making more mess we then have to compost.

Start from the roots, grow from there, please.

The #dotcons, #mainstreaming, and Build to Walk Away

Three years ago I was trying to explain something simple in language liberals might actually hear. They talk about “platform capitalism.” Fine. But I’ve been calling it the #dotcons for 20 years – because that’s what it is – a con.

The last 30 years of tech hasn’t just drifted into this mess. It’s been shaped, step by step, enclosure by enclosure, into systems designed to extract value from us. What we now call the internet is, in large part, a machine built to manipulate, capture, and profit.

The old #openweb got fenced in, and most people, especially polite liberal society, went along with it. So we need to talk about the return and the problem. Now we have a shift of the #mainstreaming is flowing back toward the #openweb, that should be a good thing. But there is a problem: people don’t leave the #dotcons behind when they move, they bring the culture with them.

What we’re seeing is a flood of the same patterns – extractive behaviour, ego performance, status games. Not from one “side,” but from everywhere. The habits built inside the #dotcons don’t magically disappear just because the platform changes.

So the real issue isn’t technical, it’s cultural. If we don’t actively mediate this influx, we won’t rebuild the #openweb – we’ll just recreate the same broken systems in slightly different code.

So why do I talk so much about compost, and mess not being the problem. Mess is necessary, but only if it composts – if it breaks down into something fertile. Right now, we’re mostly just piling it higher.

This is where projects like #indymediaback and #OGB matter. They’re not perfect, but they are native to the #openweb path: grounded in trust, process, and the #4opens rather than control, branding, and capture.

The question isn’t whether #mainstreaming is good or bad. The question is: how do we hold the cultural line so that what grows is something genuinely different? Because if we don’t, the #dotcons don’t need to defeat us. We’ll blindly rebuild them ourselves.

So why do I argue we can’t just leave the #dotcons? This is where people get it wrong, every time the #dotcons tighten control – censoring, tweaking algorithms, shifting rules – the reaction is the same: leave, build the #openweb.

Yes, build the #openweb, but the idea that we should stop organizing inside the #dotcons right now? That’s a trap, because billions of people are still there. The conversations, the communities, the movements, they haven’t magically migrated. Walking away doesn’t free those people, it abandons them, leaving the space to be shaped entirely by the #deathcult and the forces already in control.

This is #nothingnew. The #dotcons are #closedweb infrastructure. They serve power because they were built to serve power. Expecting anything else is misunderstanding the system. The real question has never been: are these platforms good? It’s: what do we do, given that this is where people are?

The #geekproblem and the exodus fantasy, is a persistent fantasy – a classic #geekproblem – that if we just build better tools, people will come. They won’t, not on their own. A clean exodus to the #fediverse or any #openweb space doesn’t happen because we post about it. Movement-building has never worked like that, people move through relationships, trust, and shared struggle – not technical superiority.

So if you abandon the spaces where people already are, you cut those pathways. The #OMN approach has always been simple to use the #dotcons as a bridge, not a home, seed organizing where people already are while focusing energy on building the #openweb in parallel to clearly keep your foundations in the #4opens.

This isn’t about purity, it’s about effectiveness, don’t fall into #stupidindividualism, the idea that personal withdrawal is more important than collective reach. This is about infrastructure and grounding, if the #dotcons can switch you off at any moment, they cannot be your foundation.

That’s why we need:

indymediaback as publishing roots
activitypub and the #fediverse as distributed infrastructure
OMN as a bridge between cultures and spaces

This is the practical expression of the #4opens: not just open code, but open process and open trust. Don’t build your house on someone else’s land, but don’t stop talking to the people still living there either. Stay in the fight, when the #dotcons clamp down, it’s not a surprise, it’s a signal of what they are, and what they’ve always been.

The answer isn’t to run away, it’s to root ourselves somewhere that can’t be shut down, while continuing to show up where the people are.

Build the #openweb, stay in the fight, keep it simple #KISS

Toward Healthier Digital Public Spaces: A Cultural and Structural Challenge

From a practical perspective, the challenges in digital and social technology are not technical, they are cultural. This may seem obvious, but it is often overlooked in policy and implementation. At a basic level, there are two broad approaches to handling disagreement and complexity in online spaces:

Exclusion-focused approaches (e.g. blocking, filtering, silencing), which reduce immediate friction but reinforce fragmentation and polarisation.

Engagement approaches (e.g. dialogue, questioning, listening, and iterative response), which are more demanding but can, over time, reduce conflict and build shared understanding.

In the current mess – shaped by strong norms of individualism and personal optimisation – the first dysfunctional approach dominates. This grows increasingly fragmented discourse, where communities become isolated and less resilient.

Understanding this we can start to show the limits of “Common Sense” in today’s mess, were governance relies on this “common sense.” Over the past four decades, economic and cultural frameworks – market-driven individualism – has controlled how we design and use digital systems contributing to:

Increased social fragmentation

Growing economic inequality

Incentive structures prioritising engagement over well-being

Environmental and social externalities (including visible #climatechaos impacts)

These outcomes tell us that existing models are not sufficient for building sustainable digital public spaces, so we need to #KISS revisit platform dynamics and structural Incentives.

The dominant digital platforms (the #dotcons) operate on business models that prioritise data extraction, engagement metrics, and advertising revenue. These incentives shape what information is amplified, how users interact and which behaviours are rewarded. While these systems are effective at scaling control, they are not in any way aligned with public interest outcomes such as trust, accountability, or democratic participation.

Current trends – ranging from disinformation to polarisation and environmental stress – highlight the limits of systems based purely on competitive, self-interested models. At the same time, alternative approaches – such as the #openweb and federated systems – offer more aligned values but face basic challenges of coordination, usability, and governance. So we need to move from fragmentation to constructive engagement, to reframe the problem, from crisis to stewardship.

This more sustainable approach emphasises stewardship over extraction, collaboration over isolation to help build resilience over short-term optimisation. This does not mean abandoning innovation or individual freedom, but rather #KISS balancing these with responsibility for shared outcomes. As the current challenges in digital spaces are not only the result of “bad actors” or isolated failures. In simple terms we need to move from systems that amplify division toward systems that support understanding and the common good.

To compost this mess we need a willingness to engage with complexity and a commitment to building systems (technical and social) that prioritise long-term public value over short-term individual gains. This is not easy work – but it is necessary if digital infrastructure is to support healthy, democratic societies.

KISS – Keep it simple, sustainable, and focused on the common good.

EU taking the open web path?

When thinking about supporting the #Fediverse, it’s important to understand that the current codebase leadership model is closer to an aristocracy than democracy. This isn’t unusual – most open source projects work this way. A small group of core developers, often a “benevolent dictator,” make key decisions. This model can work up to a point: it enables speed, coherence, and technical direction. But it does not scale in any way easily into a broader public infrastructure.

From the outside, the #Fediverse can look chaotic, and in many ways it is. A useful metaphor is an elephant stampede with people throwing paper planes at each other. That messiness is not a failure; it’s a reflection of a genuinely distributed system.

Democracy is inherently messy.
Bureaucracy, by contrast, is tidy.

The tension between those two is at the heart of the challenge outreaching to public institutions faces. Traditional organisational models – especially those involving funding – tend to concentrate power. Once money and status enter the system, decision-making quickly becomes a focus of competition and control.

We have seen this repeatedly over the last 20 years: projects that begin open and collaborative gradually centralise, and in doing so lose the qualities that made them valuable in the first place. If the goal is to support a native Fediverse ecosystem, this pattern needs to be consciously avoided.

This raises a governance question. How are decisions made once resources, funding, and institutional recognition enter the space? Without deliberate design, the default outcome is oligarchy – small groups making decisions on behalf of many. This is not only a moral failure; it is a structural tendency of complex organisations.

The strength of the #Fediverse is that it is radically different from mainstream platforms. It is decentralised, diverse, and resistant to single points of control. The risk is that, in mainstreaming paths trying to support it, we unintentionally reshape it into something more familiar, and less effective. “Common sense” approaches, based on traditional institutional models, push in this direction. If European institutions want to invest meaningfully in this space, the challenge is not simply technical. It is cultural and organisational:

How to support infrastructure without centralising control

How to enable coordination without enforcing uniformity

How to fund development without creating capture points for power

The opportunity is significant, but it requires seeing and recognising that the #Fediverse works because it is different – and ensuring that support mechanisms strengthen that difference, rather than smoothing it away.

The value of the #Fediverse comes from its cultural roots in the #openweb

Beyond Blocking: Building Trust Infrastructure for the Open Web

A Policy Case for Commons-Based Moderation in the Fediverse

The problem with the current approach

The normal response to harmful content and behaviour on federated social platforms today is the block. Instance administrators block other instances. Users block other users. Communities build blocklists and share them. This is understandable – it is the tool available – but it is not a solution. It is, at best, a temporary containment strategy.

Blocking is the digital equivalent of closing the curtains. The problem does not go away. The harmful actor does not change. The tension, between open participation and community safety, between freedom of expression and protection from harm, is not resolved. It is deferred, and at a cost to the openness that makes the Fediverse worth defending in the first place.

When entire instances are blocked, legitimate users on those instances lose access to communities they value. When blocklists are the primary moderation infrastructure, the communities that maintain them acquire disproportionate power over what the network sees. The default is isolation, the Fediverse fragments, not because of any external threat, but because of its own defensive reflexes.

This matters beyond the technical community. The Fediverse represents the largest functioning alternative to corporate social media. It is, in the most literal sense, public digital infrastructure owned by nobody and available to everyone. How it handles the tension between openness and safety determines whether it can scale to serve democratic societies, or if it remains a technically interesting experiment for a self-selecting community.

The #4opens principle and why it matters for policy

The Fediverse is built on a set of principles called the #4opens: open data, open source, open standards, and open process. These are not just technical preferences, they are a statement about what public digital infrastructure should look like – transparent, accountable, forkable, improvable by anyone.

The fourth open – open process – is the most politically significant and the most underdeveloped. It means that governing our communities, including how we handle conflict and harm, should be visible, contestable, and collectively grown. Not handed down by a platform’s trust and safety team or enforced by an opaque algorithm, not dependent on the goodwill of a instance administrator.

The current state of Fediverse moderation largely fails this test. Moderation decisions are made locally, inconsistently, and without shared infrastructure for collective reasoning. The result on balance is less freedom, more a patchwork of micro-kingdoms, each with its own rules, enforced by blocking the kingdoms whose rules they disagree with. This is not a stable foundation for the kind of digital public sphere that European democratic values require.

The commercial platforms are not the solution – but they are in the room talking loudly

Commercial social media platforms – what we call the #dotcons, shorthand for the dot-com era corporations that monetised public digital space, are present in or adjacent to the Fediverse. Meta’s Threads now implements ActivityPub, the protocol underlying Fediverse federation. This means that the same open standard that allows community-run instances to talk to each other also allows a platform with three billion users and an advertising-driven engagement model to participate in the same network.

The response in parts of the Fediverse community has been, predictably, to block Threads at the instance level. This is coherent as a local decision. As a strategy for the #openweb, it is kinda self-defeating. Blocking Meta does not make Meta go away, it does not change Meta’s incentives. It does not protect users who remain on Meta from the harms of algorithmic amplification. And it does little to build the alternative infrastructure that would give those users somewhere better to go.

The principled response to commercial platform encroachment on the openweb is not isolation, it is to build commons infrastructure so robust, so trustworthy, and so genuinely useful that the value proposition of centralised platforms diminishes. That means solving the moderation problem properly, not routing around it.

What trust-based flows offer that blocking cannot

The research and development work at projects like the Open Media Network (#OMN) points toward a different model: moderation not as exclusion but as flow management. In a trust-based flow architecture, content does not move through the network based on algorithms optimising for engagement, nor is it blocked at the border by administrators making binary decisions. Instead, it flows – or slows, or stops – based on trust relationships that communities build and maintain themselves. Trust is local, it is composable, different communities will apply different trust filters to the same content without requiring global consensus or any centralised authority.

This model has several properties that should interest European policymakers directly:

Accountability without centralisation. Trust relationships are explicit and auditable. When a community decides not to propagate certain content, that decision is visible and contestable within the community. This is categorically different from both corporate content moderation (opaque, unaccountable) and simple blocking (binary, irreversible).

Resilience against capture. Because trust is distributed and local, there is no single chokepoint that a bad actor – commercial, state, or otherwise – can capture to control information flows across the network. This is critical infrastructure resilience in the same sense that distributed energy grids are resilient against single points of failure.

Reversibility. The rollback function – the ability to re-evaluate historical content visibility based on updated trust relationships – is something no current platform offers at scale. It means that moderation decisions can evolve as communities learn, rather than being permanently encoded in block lists that few people maintain.

Scalability without hierarchy. Top-down moderation breaks down as communities grow. Moderators experience burnout and trauma. Rules based decision making become inconsistent. The trust-based model scales horizontally – as the network grows, the trust infrastructure grows with it, because it is built into the relationships between nodes rather than concentrated in any central authority.

The culture question is not separate from the technical question

It is a constant mistake to read this as purely a technical, you cannot build a healthy online culture without infrastructure – and you cannot build the working infrastructure without clear visions of what culture looks like.

The Fediverse community is currently navigating this without adequate tools. The result is a recurring cycle: a wave of new users arrives, often fleeing a crisis on commercial platforms. The existing community debates how to handle them. Blocking becomes the instrument of cultural negotiation. Fragmentation follows. The cycle repeats.

What is needed is not better blacklists, we need infrastructure that makes constructive engagement the path of least resistance, where trust can be extended incrementally, withdrawn proportionally, and rebuilt over time. Where communities are not forced to choose between openness and safety because the tools exist to manage both simultaneously.

This is a social and political problem that has a technical component. The Open Media Network project is one concrete path to solving it, building on mature existing infrastructure, proven open standards, and a decade of practical experience in grassroots on the ground and online federated media.

What European public investment can achieve here

European public funding for digital commons infrastructure has a strong track record. The NGI Zero programme has supported foundational work on everything from secure routing protocols to private messaging to federated video platforms. This investment compounds: open source outputs are reused, extended, and built upon by developers and institutions across the continent and beyond.

The case for investing in trust-based moderation infrastructure for the Fediverse is straightforward as the problem is real, well-documented, and getting worse. The Fediverse is growing, but without better tools for managing harmful content and building coherent information flows, its growth will hit a ceiling defined by the limits of volunteer moderator capacity and the inadequacy of binary blocking as a governance tool.

The solution is technically tractable, the components exist, the protocol exists, the codebases exist, the community exists. What is missing is the focused R&D investment to implement trust-based flows as working, deployable, open infrastructure.

The alternative is worse, if the Fediverse fails to solve this problem, the vacuum will be filled either by commercial platforms extending their reach into the federated space on their own terms, or by the continued fragmentation of the #openweb into isolated communities talking only to themselves. Neither outcome serves European democratic values or European digital sovereignty.

The investment required is modest. The upside, is a functioning commons layer for federated media distribution, owned by nobody, available to everyone, accountable to the communities it serves, is large. The time to build it is now, before the structural problems of the current moment ossify into the permanent architecture of the next generation internet.

Why you should help

The internet’s public square is privatised, algorithmically controlled for “engagement” over any idea of truth, and placed under the control of a handful of American corporations with no accountability to European citizens or values. The #Fediverse is the most credible existing alternative – but it lacks the shared infrastructure to function as a native commons for news and media. #OMN builds that infrastructure: trust-based, community-controlled, transparent, reversible, and owned by nobody. At €45,000 for a proof of concept, it is one of the cheapest possible investments in the long-term health of European digital public life. If it works – and the technical and social groundwork suggests it will – it becomes the plumbing for a Fediverse that can actually be used to serve democratic societies rather than more #techshit alongside the current #dotcons platforms that undermine them.

Why this matters – because the #WWW was stolen – Designed as commons at CERN, decentralised, open, nobody in charge. What we have today is instead is five American corporations controlling the information diet of billions of people. Facebook decides what news you see. YouTube’s algorithm decides which voices get amplified. Twitter/X decides who gets banned. None of these decisions are transparent, accountable, or reversible. They are made by private entities in pursuit of control, advertising revenue and engagement metrics – not truth, not public interest, not democracy.

The #Fediverse exists as a rejection of this, it’s the largest real functioning alternative to corporate social media, with millions of people on thousands of servers, federated together, nobody owning the whole thing. It works. It’s growing. But it has a weakness: it’s kinda fragmented at the commons layer. There’s no shared infrastructure for how news and media actually flows across the network in any trustworthy and coherent way.

That’s the gap #OMN fills, but why? Most people don’t think about internet infrastructure. They think about whether they can trust what they read. Whether the news they see is real. Whether the platform they’re on is working for them or selling them. Whether they can do anything when something goes wrong.

Right now the answer to all of those is: it depends entirely on decisions made by people you’ll never meet, for reasons you’ll never know. OMN proposes something different. If your community trusts a source, a trust flow, you see it. If they don’t, you don’t. And that decision is yours, reversible, transparent, locally controlled.

For a journalist in a small country trying to get independent news out, this is the difference between having infrastructure that works for them and being at the mercy of a platform that can deplatform them overnight. For a community archive trying to keep historical memory alive and accessible, this is the difference between dependence on Google’s goodwill and owning your own distribution. For an ordinary person trying to figure out what’s true, this is the difference between an algorithm designed to maximise your outrage and a network shaped by people you actually trust.

Bureaucracies fund things slowly, in ways that often serve existing power structures rather than challenging them. But digital sovereignty is an existential European concern. The EU has spent years trying to regulate American platforms – GDPR, the Digital Services Act, the Digital Markets Act – and the platforms have responded with compliance theatre, token gestures, and armies of lawyers. Regulation of concentrated private power is a losing path. The only actual answer is to build the alternative infrastructure so that people have somewhere else to go. That’s what the NGI Commons Fund is for and what #OMN does.

The EU should not only be funding products, it needs to fund commons infrastructure – the plumbing that nobody owns and everyone can use. Like funding roads rather than funding a logistics company. The outputs are open source, meaning any European media organisation, any local community, any public institution can pick this up and use it. No lock-in. No dependency on a vendor who will be acquired or shut down.

It’s cheap, with the second stage scaling across Europe with institutional partners, building on European strengths. The Fediverse is disproportionately European. Mastodon was built by a German developer. The culture of digital commons, open standards, and public interest technology is stronger in Europe than anywhere else. This project is native to that tradition. It’s not asking Europe to compete with Silicon Valley on Silicon Valley’s terms – it’s asking Europe to build the alternative on its own terms.

The problem #OMN solves is getting worse, not better. Disinformation, algorithmic radicalisation, platform capture of public discourse – these are not abstract threats. They are actively destabilising European democracies. Funding the technical infrastructure for trustworthy, community-controlled information flows is not a nice-to-have. It is digital public health infrastructure.

#KISS

Thematic call: NGI Zero Commons Fund

Organisation: Open Media Network (unincorporated community project, fiscal hosting in Belgium via OpenCollective) Country: United Kingdom General Project Information Proposal name: Trust-Based Media Flows for the Fediverse (#OMN) Website / wiki: https://unite.openworlds.info/Open-Media-Network/Open-Media-Network

Abstract

Can you explain the whole project and its expected outcome(s)?

The Open Media Network (#OMN) is a protocol-driven, federated media infrastructure built on top of ActivityPub and the Emissary codebase (emissary.dev). It addresses a real gap in the current Fediverse: while platforms like Mastodon, PeerTube, and Lemmy are federated at the instance level, there is little coherent cross-platform layer for trust-based content flows, moderation, or news aggregation. Each instance operates largely as its own silo, moderation is hierarchical and per-server, and there is no shared commons model for media distribution across the ecosystem. #OMN proposes a minimal, compostable interaction model – the Five Functions (#5F): Publish, Subscribe, Moderate, Rollback, and Edit Metadata – implemented as a flow layer on top of existing Fediverse infrastructure. Content moves through the network as objects flowing through pipes and holding tanks, filtered and shaped by trust relationships between nodes rather than by opaque algorithms or centralised authority.

The central R&D question is: can trust-based moderation and distribution flows replace algorithmic amplification in a federated news ecosystem? Expected outcomes of this first-stage grant: By Month 3: A technical specification of the flow architecture; a prototype flow service routing ActivityPub objects between two instances; documentation of existing Fediverse flow patterns; early integration with one platform (likely PeerTube). By Month 6: A cross-platform prototype connecting at least two Fediverse systems; a working demonstration of trust-based moderation flows; a public code repository and documentation; and a user-facing prototype via the #makinghistory test environment (https://unite.openworlds.info/Open-Media-Network/MakingHistory). All outputs will be released under recognised open source licences. The project follows the #4opens framework: open data, open source, open standards, and open process.

Have you been involved with projects or organisations relevant to this project before?

Yes. The project lead, Hamish Campbell, has over 40 years of experience in grassroots media and technology, including early involvement with Indymedia – the pioneering open publishing news network – and more than 8 years working directly with the Fediverse and ActivityPub community. The #OMN conceptual framework has been developed over this time and is documented extensively in the project wiki, SocialHub, and at https://hamishcampbell.com. Developer Michael has contributed to #OMN concepts and logic for 10 years and is currently building the #makinghistory reference implementation. Ben, the core developer of Emissary, brings specific expertise in the codebase that will form the technical foundation of the project. Alex brings potential DAT/distributed storage support, and IKA will work on testing and rollout.

Requested Support Requested Amount: €45,000

Explain what the requested budget will be used for. Does the project have other funding sources, both past and present? A breakdown in the main tasks with associated effort is appreciated. Make rates explicit. The budget covers a lean, seed-stage proof of concept with no prior external funding. There are currently no other funding sources. The budget breakdown can be found in the attached PDF (funding). Roles: Hamish Campbell (project lead, coordination, documentation, community engagement) and Michael Saunders (primary development, UX, system logic). Additional contributors (Ben, Alex, IKA) are contributing on a voluntary/community basis during this seed phase. Work packages and approximate effort: WP1 Research & Specification (Months 1–2, ~25% of effort): Architecture design, gap analysis of existing Fediverse tools and flows (PeerTube, Lemmy, Mastodon), and documentation of trust-flow patterns. Output: Technical design document. WP2 Core Development (Months 2–5, ~45% of effort): Flow service implementation on top of Emissary; ActivityPub integration for the #5F model; and a trust-based moderation layer extending Emissary’s existing block/flag capabilities. Output: Working prototype codebase. WP3 UX & Prototype (Months 3–5, ~20% of effort): #makinghistory user interface; dual-layer UX (simple and advanced modes); and WCAG 2.1 accessibility compliance. Output: Testable user prototype. WP4 Testing & Documentation (Months 5–6, ~10% of effort): Community testing and iteration; public documentation and reports; and an open knowledge base of what works and what fails. Output: Public documentation, reports, and reusable design patterns. LINK PDF and wiki

Compare your own project with existing or historical efforts.

The closest existing efforts are: Mastodon’s built-in moderation tools: per-instance block lists and the Fediblock community blocklist. These are instance-level tools – they do not create cross-platform trust flows or shared content aggregation. #OMN operates at the network layer, not the instance layer. Fediseer: a trust registry allowing instances to vouch for each other. Fediseer addresses instance-level reputation but does not implement content flow logic, rollback, or metadata editing as network functions. #OMN builds a compostable flow model on top of the kind of trust signals that Fediseer represents. GNU Social / Friendica: older federated social platforms with some aggregation capability. These predate ActivityPub’s consolidation as the dominant standard and do not address the cross-platform news/media commons use case. Indymedia (1999–2010s): the historical precedent for open publishing federated media. Within the wider project, #OMN explicitly revives and modernises the Indymedia model for the ActivityPub era via the #indymediaback reference implementation, addressing the unfinished work of that tradition. The #makinghistory project grows from, and shares, this same established workflow. Bonfire networks: likely related, but unclear in scope and function. Attempts to install and use it have not clarified its approach. It may be trying to address similar problems, but this remains uncertain. The key difference of #OMN: it is not building a new platform. It is building a protocol-level flow layer that works across existing Fediverse platforms, implementing trust-based content propagation as commons infrastructure rather than as a product. See included PDFs.

What are significant technical challenges you expect to solve during the project?

Trust flow implementation: Designing and implementing a data model for trust relationships between federated nodes that is lightweight, compostable, and expressible via or alongside ActivityPub. Trust is local and subjective – the system must allow different communities to apply different trust filters to the same content flow without requiring global consensus.
Rollback across federated state: Implementing the rollback function (re-evaluating and reshaping historical content visibility) in a distributed system where content has already propagated to multiple nodes. This requires a time-aware, local re-indexing approach rather than a global delete mechanism.
Cross-platform content normalisation: Aggregating content objects from Mastodon (short-form social), PeerTube (video), and Lemmy (forum) into a common JSON-LD content model with a consistent trust trail, despite these platforms having different ActivityPub implementations and object schemas.
Search actors as push feeds: Implementing the “content finds you” model – where a defined search query becomes a persistent ActivityPub actor that pushes matching new content to subscribers – requires extending Emissary’s existing subscribable search engine capability.

Describe the ecosystem of the project, and how you will engage with relevant actors and promote the outcomes.

The primary ecosystem is the Fediverse: the network of federated, open-source social platforms running ActivityPub, including Mastodon, PeerTube, Lemmy, Friendica, and many others. This ecosystem has grown substantially (estimated 10+ million active users across thousands of instances) but remains technically fragmented at the commons/media layer. The project builds directly on the Emissary codebase (https://emissary.dev), an existing ActivityPub-native Go application. Engagement with the Emissary community is embedded in the team through Ben’s mentoring role.

Wider ecosystem engagement:

The project will contribute design patterns and documentation back to the broader Fediverse developer community via public code repositories, the project wiki, and events. The #makinghistory test phase connects us to existing archives such as Bishupsgate, Maydyroom, the Peace Museum, and the Campbell Family Archive, providing access to extensive datasets as well as outreach to their administrators and users. The five community events included in the budget are specifically designed to recruit contributors, gather real-world feedback, and expand the network of participating nodes.

Promotion of outcomes:

Outcomes will be shared through the Fediverse itself (maintaining an active presence on ActivityPub-native platforms and legacy social media), via open-licensed documentation, and through NGI/NLnet networks and events. This first-stage grant is explicitly designed as a seed and proof-of-concept phase, with a larger second-stage proposal planned to deliver a fully production-ready system once the core architecture is validated.

See attached PDFs.
Powered by Forgejo

Technical Specification Download

Answering the small print for funding Download

Funding Download

Would like to thank all the meany people who helped with this.

What happened over the last ten years on our Fediverse path

The first steps were good. #Socialhub emerged as a genuinely grassroots space, shaped to maintain the integrity of the #activertypub native reboot. It grew directly out of the #activertypub affinity group itself – rooted in lived practice rather than imposed structure.

So what motivated this native path? The current #openweb reboot wasn’t exactly planned – it was, in many ways, serendipitous. During the #WC3 process, the usual mainstream players were largely absent. That gap created space for an alternative cohort to step in and shape things in a more “native” way. This is rare. Normally, these processes are dominated by institutional and corporate interests, but for a moment, we had something different – and it worked.

From that strong beginning, #Socialhub grew into a real, functioning community. Its high point was during the Fediverse outreach to the EU, when there was a sense of shared purpose and direction. The social and technical sides were in balance, and the space felt alive, open, and productive. But over time, things shifted.

The rapid growth of the Fediverse brought in many people without any grounding in “native” #openweb culture. The influx – particularly from Twitter – changed the tone and priorities. This wasn’t entirely negative; growth always brings energy and diversity. But it also brought confusion, and a drift away from the original focus.

At the same time, there was a strong, increasingly dogmatic shift toward the technical side of #activertypub, at the expense of the social layer that made it meaningful. The balance tipped. The core crew thinned out, and newer, more tech-focused contributors filled the space. This mirrored the rebooting of the #WC3 process, and the two together created a difficult, often unspoken tension over direction and responsibility. Governance also became an issue. The line:

“To use the forum, you must agree to these terms with Petites Singularités, the company that runs the forum.”

Made visible something that had been quietly present for a while: this was not, in practice, a community-owned space. It had an owner, with an agenda. What had been presented as a shared, grassroots commons was, structurally, something else?

This marks a deeper shift – from serendipitous emergence to more deliberate control.

A short update: how we are failing

We didn’t fail because of bad intent. We fail because we didn’t hold onto the balance that made the space work.

We allowed the social layer to be sidelined by the technical.
We didn’t build clear, native governance while we still had the chance.
We mistook growth for success, without mediating the cultural shift it brought.
We let ownership and control consolidate quietly, instead of addressing it openly.
And when tensions emerged, we defaulted to avoidance and #BLOCKING, rather than doing the messy work of resolution.

In short, we lost the thread of the #openweb path by not actively maintaining it.

Where that leaves us now? We are now in a more complex, more conflicted space. The community is bigger, but less coherent. The vision is more diluted, but still present, if we choose to pick it up again.

The solution isn’t simple. It likely involves some form of real, lived democracy, and a return to explicitly valuing the social processes alongside the technical ones. And maybe the only solid ground we still have is this: Grassroots is always messy, that mess isn’t a flaw – it’s how you know it’s real. The challenge is not to remove the mess, but to hold it together well enough that it can still grow.

The value of the #Fediverse comes from its cultural roots in the #openweb

The Digital Commons: The Ground We Already Stand On

So how do we mediate this fear?

Who would have though this would sum up our needed path for the #Oxfordboaters and the #fedivers?

You would have to be an #asshole to unthinkingly disagree with what we are doing and pretty wise to thinkingly disagree with the path. Which one are you? So why are we in such a mess? Because people are acting from fear. Not always consciously, not always honestly – but fear is the driver.

Fear of losing control.
Fear of losing status.
Fear of uncertainty.
Fear of each other.

And when fear leads, people grasp for control. They close things down, centralise, gatekeep, and default to the safe, known paths of the #closedweb and institutional power. That’s how we get the current mess – top-down structures trying to manage what was meant to be lived, messy, and shared.

In #OMN terms, this isn’t a technical failure, it’s a cultural one. A failure to hold open processes in the face of discomfort. So how do we mediate this fear?

Not by pretending it isn’t there. And not by fighting it head-on – that just feeds it. We mediate fear by building trust through practice:

Keep things open (#4opens): transparency reduces fear of hidden agendas. When people can see what’s happening, they relax.

Lower the stakes: small, reversible steps instead of big, risky commitments. Let people edge in rather than jump.

Normalise mess: show that not everything has to be controlled to work. Messy, lived processes are not failure, they’re how real communities function.

Create shared doing: fear shrinks when people work together on tangible tasks. Composting, media, infrastructure – doing builds trust faster than talking.

Hold both fluffy and spiky: the fluffy path makes space for people to come in; the spiky path protects that space from being captured or hollowed out. You need both, visibly and honestly.

Refuse false clarity: the #dotcons sell certainty and simplicity. The #openweb is different, it’s about holding complexity without collapsing into control.

And maybe most importantly, stay present. Fear thrives in abstraction, it weakens in lived, grounded relationships. In the end, mediating fear isn’t about convincing people with arguments. It’s about creating environments where fear has less reason to exist.

This is the same dynamic you can see with Oxford boaters. The river culture is native, messy, negotiated, based on lived practice and mutual understanding. People want the freedom to move, to live lightly, and not be bound by rigid landlord rules. But when outside structures push in – formal control, ownership models, enforcement – they reshape that culture into something else. The tension isn’t really about rules or functions; it’s about which culture gets to define the space.

That’s the real work of #OMN: not only building tools, but growing the social soil where people feel able to act without retreating into control.

Individual fears

How things can change

Hope this helps compost some of the mess building up. It’s something we all need to do and have responsibility for.

Groups don’t usually fail because of external pressure, they fail because they turn inward and burn energy on themselves. If you want a calm, #KISS path that actually holds diversity without collapsing, we need a few simple lived – traditions and mythos – not heavy governance, not ideology battles, just grounded #KISS practice:

Keep the core action very small and clear – a shared purpose. If people can’t easily answer “what are we doing?”, drift and conflict creep in.

In twine “doing” with “talking” Most infighting comes from too much abstract discussion. Doing space – Talking space – Don’t let one swamp the other.

Protect focus like it’s fragile (because it is), the biggest risk isn’t disagreement – it’s distraction. When things start spiralling bring it back to “what are we building this week?” if it doesn’t help, park it

Default to trust, but design for friction, diversity is strength. But unbalanced diversity = chaos. So let people approach things differently, but require shared outputs – If it doesn’t produce something, it doesn’t dominate attention.

No purity tests, this is where diversity dies. People will come with different politics, paths (fluffy vs spiky) and have different priorities, that’s fine – as long as they don’t block others doing the work.

Make conflict low-energy, not zero-conflict – we won’t avoid disagreements. Trying repression = explosion later. Instead, keep arguments short, move unresolved tension into parallel paths (“try both”) and let results decide, not personalities. This is the “compost” approach we need to talk about – don’t fight the mess, process it.

Grow by doing, not convincing, you don’t need everyone to agree. You need visible, working examples. Let people see it working – that’s what grows a community of action.

What we are sketching and building is the hard middle path of not rigid control (kills growth) and not total openness (creates chaos). But a light structure that keeps things moving.

#Oxford example:
Oxford boaters are sovereign, keeping the free-flowing life of the river, not bound by the old rules of the landlords.

#Fediverse example:
The Fediverse is native to the #openweb path. We judge by the #4opens, and walk with power.