This is the seventeenth conversation of the 100+ Conversations to Inspire Our New Direction (#OKFN100) project.
Starting in 2023, we are meeting with over 100 people to discuss the future of open knowledge, shaped by a diverse set of visions from artists, activists, academics, archivists, thinkers, policymakers, data scientists, educators, and community leaders from around the world.
How can openness accelerate and strengthen struggles against the complex challenges of our time? This is the key question behind conversations like the one you can read below.
*
Today’s conversation is with Jaya Klara Brekke, CSO at Nym Technologies, a decentralised privacy system combining a mixnet with a blockchain node reputation and reward system.
Jaya has been working as creator, designer and curator on projects related to the political economies of infrastructures for the past ten years. She holds a PhD from Durham University Geography department and was an expert advisor to the European Commission. Her research contributes to public, industry and policy debates about emerging data economies, decentralised data infrastructures and blockchain technology internationally.
Her past projects include research and development on ethics for B9Lab blockchain developer training company; Crisis-scapes, on the impact of the financial crisis on public space in Athens; D-CENT, a European-wide project for the development of open citizen engagement technologies; and Flesh & Concrete, a research project and art programming on the effects of the construction of the supervia poniente highway in Mexico city.
This conversation with Ksenia Ermoshina, senior researcher at the Center for Internet and Society (CIS/CNRS), took place on 9 July after her keynote ‘Another Network Is Possible’ at The Tech People Want Summit (video below), moderated by Lucas Pretti, OKFN’s Communications & Advocacy Director.
We hope you enjoy reading it.
*
Ksenia Ermoshina: I just finished a podcast about Narodnaya Volya, a Russian organisation that was one of the first decentralised groups to organise against the Empire at the end of the 19th century. They used a bunch of very complicated tools to maintain their organisation and carry out direct actions without being discovered, despite not having the internet.
I’m drawing a comparison between these crazy people from the 19th century and the Nym group because they both use social and technical means of obfuscation. In my view, this is a great achievement and an interesting design choice by Nym. It also reminds me of those old-school anarchist techniques where people pretended to be a polite, noble couple who owned a small store. In reality, they were making dynamite to fight corrupt police officers. This example of combining humans and technology persists.
Building on that, I’m curious about how you enrol operators and motivate these meme squads. I guess financial motivation is not the only factor. It’s important, of course, but what else motivates these people? What are the main ideas or values that people talk to you about when explaining why they want to be in a squad?
Jaya Klara Brekke: That’s a really good question! It’s an interesting aspect of Nym. Our operator community is extremely diverse. Some of our operators also run tour relays. They really believe in operating this kind of privacy infrastructure and are driven by their values. So I would say that’s one type of operator who is more an old-school hacktivist.
Then there are more profit-motivated Web3 infrastructure operators. These are the crypto bros. They run validators and infrastructure for lots of other Web3 projects and are primarily motivated by the rewards.
From our perspective though, one of the things I really like about Nym is that it manages to bring these different communities together, all with the shared function of enabling a strong infrastructure for end users. Whether you’re there for the money, because you believe in privacy or because you’re an activist squad that wants a secure gateway, these are all valid reasons.
The end result we’re interested in is reliable infrastructure that users can make use of. Nym is a company, an ecosystem and a community. We span wildly different contexts, both culturally and in terms of the regions of the world we operate in.
I think it remains to be seen, but I believe this will be a strength as we move forward.
Ksenia Ermoshina: Thanks a lot! Now, I have one technical question. What can a node operator see when they are running a node? What information can they access about other people’s traffic?
Jaya Klara Brekke: Well, you can’t really see anything. The way it works is that you have this type of onion encryption, right? As an operator, you receive a packet. You unwrap one layer of the packet, which tells you where to send it next. That’s the reason why we have five layers in the actual network.
I didn’t go into detail on that diagram, but effectively there’s a gateway and three layers of mixed noise that mix traffic and add dummy traffic, as well as an exit gateway. The reason we have exactly five layers is so that there’s never a point in the network where an operator can see both the sender and the receiver.
All the operators know is where the packet came from and where to send it next. This is an important part of the design. It’s also how the stinks packets are structured. It’s a layered encryption format. As an operator, you receive the packet, strip one layer down and see where to send it next, and that’s it.
Ksenia Ermoshina: Great! That is really reassuring. One of the current focuses of my research and practice is on remote, stable network conditions in areas with precarious connectivity, as well as on how some secure tools fail due to accessibility and network instability issues.
Have you conducted any trials, such as testing Nym in areas with strict censorship or networks that are unreliable due to natural disruptions? How do you approach the problem of accessibility and resilience?
Jaya Klara Brekke: That’s a really important question.
When it comes to reliability in low-connectivity areas, it’s a new research strand that we still need to work on. We’re currently further along in our efforts around censorship resistance and the strategies we’re planning to deploy.
We’ve had people testing Nym in places like Russia, China, and Iran. The mixnet itself – the noise-generating mixnet – has proven more successful in censored regions. Nym has two operational modes: one using the mixnet and another using a simpler two-hop relay network. The two-hop relay tends to be more susceptible to censorship, while the mixnet is more robust.
So far, we’ve seen people in Russia manage to use it depending on their ISP. We’ve also had some success in China. Iran has been more difficult, especially since the authorities took down much of the network infrastructure recently. There have been moments of success, but overall it’s unreliable.
We’re working on more sophisticated censorship-resistant strategies that should roll out over the next six months. It’s very much a cat-and-mouse game. We’re also looking into protocols like Amnesia, W-GAR, and QUIC to improve resilience.
Ksenia Ermoshina: I’m really interested in this censorship cat-and-mouse game and how we might escape it. I’ve been thinking about mixnets operating within subnets – essentially building a splinter net. Iran’s recent activation of their national intranet between June 17 and June 25 showed just how capable their regulators are at isolating the country.
Having squads within territories that frequently experience shutdowns could at least provide local support. Russia is a good example. It’s a country with high technological skills and a lot of infrastructure. Many developers there are still active and highly skilled, and the infrastructure is relatively cheap. So I wonder if we could provide privacy layers inside isolated networks. Russia’s DPI, as mentioned in the chat, can inspect some traffic, but I believe it wouldn’t do much against a well-implemented mixnet.
Jaya Klara Brekke: Absolutely. I think collaboration with people inside Russia would be very valuable.
People living under censorship conditions often have exceptional technical skills – not just through formal training, but from necessity. We’ve found great collaborators in Iran and China as well.
Ksenia Ermoshina: A question from the chat: Would the noise-generating mixnet increase bandwidth requirements, since the packets are larger? If so, how much additional data does this introduce? Are we talking about doubling, tripling the bandwidth?
Jaya Klara Brekke: It varies. The amount of cover traffic can differ significantly, and we’re working on tuning it to make it as efficient as possible without using unnecessary bandwidth.
You’re right – at the moment, the system uses extra bandwidth and also drains more battery. It doesn’t come for free. But with better parameterisation and smarter censorship resistance – like shaping packets to camouflage them instead of relying on dummy traffic – we can optimise performance.
We’re still at the early stages. Future iterations will improve efficiency, including adapting packet sizes for different use cases. We’ve realised how specific everything is to the context. Some types of data have strong digital fingerprints and require a lot of cover traffic to obscure, while others do not. So we’re working on these details.
Ksenia Ermoshina: Another question from the chat: What are the trade-offs of using Nym compared to other approaches?
Jaya Klara Brekke: Many of the trade-offs we’ve discussed already – particularly the extra bandwidth and increased latency due to cover traffic. These are ongoing areas of research and optimisation. We’re exploring new encryption methods to reduce cryptographic overhead and new traffic-shaping strategies to improve speed.
Latency, bandwidth, and speed are the main trade-offs. But they also present fascinating engineering challenges. We welcome more people to look into them, publish papers, and help us move forward.
Ksenia Ermoshina: There’s another question in the chat: What’s Nym’s business model? Some people are skeptical, especially when headlines mention $300 million. Can you clarify?
Jaya Klara Brekke: Good question. No, we do not have $300 million. That number probably comes from media coverage of a fund we helped launch a few years ago.
Here’s the background: as Nym became better known, we received investment from venture capital firms and did a token launch. The funds raised came from both VC firms (all listed on our site) and the general public. That’s how we financed the early development.
We also received early-stage funding from the European Commission for research on the mixnet. The $300 million figure refers to a VC fund we helped raise, where various firms pledged funding for privacy tech startups we would refer to them. That was not money in our treasury – it was a promise to fund others.
As for our business model: we used the initial funding to build the network and app. We’re now transitioning into a paid service to extend our runway. We’ll continue applying for research grants and funding for R&D.
Ksenia Ermoshina: Thank you. This idea of “the tech people want” is really interesting. There’s a desire embedded in it. You mentioned the shift in communication – from warm, private family photos to shouting on public platforms like Twitter. These two paradigms differ greatly.
I’ve observed a migration from the open, global-village internet towards smaller, safer islands of communication. Mastodon is a great example – a kind of exile from the Wild West of tech bros.
Choosing smaller instances and simpler tools reflects the idea that less is more. Like John Cage, who gave us four minutes and 33 seconds of silence – he did a lot for noise music while making very little noise himself.
Of course, infrastructure is different – we need many hands and a lot of invisible labor. But we don’t need to be the biggest VPN or messaging app. That’s also the attitude we have at Delta Chat. If we can provide strong privacy for groups that trust us, that’s already a success.
We should move away from wanting to be the “one app to rule them all” and instead focus on being a stable, trusted option for those who choose us. Nym seems to be moving in that direction – where users are also providers. It’s a model I appreciate.
Jaya Klara Brekke: Well said. I almost don’t want to add anything. That feels like the perfect place to wrap up.
But just a small anecdote: I agree, there’s a general fatigue and exhaustion with big platforms – not just among privacy enthusiasts, but also ordinary users. People are shifting away from posting publicly to more private group chats – these small, trusted “archipelagos.”
The word “squad” comes from a piece by the group Other Internet called Squad Wealth. It noted this exact trend: people coming together in groups of friends. That scale is what feels right to many today.
What we’re building with Nym reflects this – offering alternatives to platforms that exploit our social lives. We want to build things that feel real again.
Lucas Pretti: Amazing conversation. I have one last question for both of you: How would you describe the tech you want?
Jaya Klara Brekke: I feel like we’ve partly answered that already. But I’ll add this: the idea of open protocols run in a decentralised way has been around since the beginning of the internet. It’s not new, but maybe we’ve come back to it with a fresh perspective.
One idea I find compelling is: “Know who runs your infrastructure.” At Nym, the node operators are visible and accountable, inspired by early cypherpunk thinking – “transparency for the powerful, privacy for the rest of us.”
Infrastructure operators are the powerful actors because users depend on them. So we want transparency around who they are. That kind of visibility could have a regenerative impact – not just technically, but socially.
Ksenia Ermoshina: I agree. I’m a fan of local-first technology and minimalism. I’ve learned from my kids the value of being offline for long stretches. It’s a privilege, yes, but it reminds me how important it is not to over-rely on devices.
In some hacker-run farms in Italy, I’ve seen shared computers – just like in old libraries. Imagine the level of community trust it takes to share a machine like that. No cryptography can protect you if your peers aren’t trustworthy.
Personally, I prefer federated protocols and open standards. That’s why I dislike platforms like Bluesky that reinvent the wheel instead of adopting existing protocols like ActivityPub.
Old, tested standards like email are underrated. That’s why I love Delta Chat – we’re upcycling SMTP and IMAP, making them usable and beautiful again. Like upcycling furniture or recycling plastic, we should treat older protocols with the same care.
Less computers, more humans. And more standards. That’s my vision.
