# Is Censorship Resiliance Truly Necessary? by Mashbean || Ethereum Privacy Stack, Devconnect 2025

- Channel: [Ethereum Cypherpunk Congress](https://streameth.org/ethereum-cypherpunk-congress)
- Date: 2026-01-09
- Duration: 18:38
- Topics: web3, privacy, now, crypto, cryptography, blockchain, data, security, human right, rights, tech, technology, internet, open source, free, freedom, ethereum, hackers, ethics, cypherpunk, dev, developer, dapp, decentralization, bitcoin, computer, surveillance, cyber, peer2peer, p2p, love, solidity, zk, zero knowledge, education, academy, w3pn, privacy stack, devconnect, 2025, eps25, pse, privacy stewards of ehtereum, ethereum foundation, ef, Education
- Watch: https://streameth.org/watch/yt-8J1ANo5396M
- YouTube: https://www.youtube.com/watch?v=8J1ANo5396M

## Description

Here Mashbean asks why is so hard to build sustainable product around Censorship resiliance, 
wheter there is market demand for it. What contradictions they found along the way with building censorship resilience based organisation and how these contradictions are pushing them to find new solutions.

Ethereum Privacy Stack is a global privacy summit during Devconnect 2025 bringing together Ethereum builders, protocol maintainers, and advocates. 
Featuring Vitalik Buterin, Roger Dingledine, Andy Guzman, Polymutex, Ameen Soleimani, and 30+ speakers on 2 stages, celebrating privacy acceleration.

Ethereum Privacy Stack: 
http://eps25.web3privacy.info

Organized by 
Web3Privacy Now & Privacy Stewards of Ethereum

Web3Privacy now collective: http://web3privacy.info
Privacy Stewards of Ethereum: https://pse.dev/

## Transcript

[applause] Hello. Hello. Test. Yeah, nice. Let's see if my slide is already prepared. Okay, let's wait for some times and thank you for having me. Um, among the whole week, the slogan I like the most is this one right there. The privacy is not a crime. The privacy is not a crime. First, it's not just a slogan. It's exactly what happened in the real world. Today, I am going to share the stories from our side in Chinese speaking communities because in some authoritarian region, we are kind of terrorists, not from the democratic government, but from the other side of the world. So, my name is Mash Bin and I work as a general manager of matters.com. is a censorship resistant publishing platform for writers especially in Chinese speaking society. Before that I work in Taiwan government for digital democracy and I also the member of Soviet culture society. Okay, let's see if the slide is already on. So we are waiting for the slide but I think I can just keep going on. So today I want to talk about a simple but uncomfortable questions. If censorship resistant is so important, why is it so hard to build a sustainable product or organization around it? I will work through the three things. The first is whether there is a real market demand and how people working on this survive. And the second is what contradictions we met after running a censorship resistant platform for five years. And the third is how these contradictions are publish pushing us to explore a different kind of a solutions. So let me start from the market demand questions. So when we talk about censorship resistant and privacy enhancement technologies, we usually think about the people at the age, the independent journalist, whistleblowers uh and oh yeah and the glass roots organizers, human rights lawyers and people inside authoritarian regimes or people in democracy who are touching very sensitive topics. These people are absolutely have a strong need to protect themselves. They need anonymity or at least judamity. They need the ability to disappear, to challenge identities, to change identities, to move across borders without leaving a clear trace. They need censorship resistant tools for publishing, for money, for coordination, for voting, for archiving evidence. In theory, this is exactly what censorship resistant tools, privacy finances, private voting, and anonymous publishing platforms are designed for. But in practice, most of these people do not know these tools exist or they do not trust them or they find them too hard to use. Many of them are already overloaded just saying staying safe in the physical world expecting them to also become experts in web 3 key management, IPFS and private privacy wallet is not realistic. And this is exactly what is happening in Chinese speaking community. Not many people know about it because most of the activists they won't speak publicly or they are already diminished. So we have a strange situation. The age community has a strong moral demand for censorship resistance. But the visible demand in the market is small, fragmented and often cannot pay. On the supply side, the people who build these tools face a double risk. They take the normal startup risk of not finding a business model. At the same time, they take personal security risk because they are helping people who challenge powerful actors. If you are building a dating app, you worry about the product market fee PMF. But if you build a censorship resistant tools for activist, you worry about product product market fee and about to go into jail getting sued or being targeted by an authoritarian state or by a powerful corporations. So how do these projects survive? Traditionally many rely on philanthropies, human rights nos or grants from democratic governments. Some try to raise venture capital but censorship resistance for descendants is not exactly the classic hockey sticks market theory. Today the environment is even harder. The budgets from democracy and human rights programs are being cut by the US government these years and the regulation around crypto privacy and national securities is getting tighter. Authoritarian regimes are more aggressive and even democracy sometimes treat privacy tools as suspicious just like the last talk. This means that many censorship resistant projects live from grant to grant never sure if they will survive the next year and the people working on them often burn out or quietly move to safer more profitable industries. So the first answer is yes there is a demand but it doesn't behave like a normal market. It's closer to demand for emergency services. You don't buy a fires truck as a consumer product. You still needed to exist and somebody may be interested of keeping them exist. Now let me talk about what we actually learned from the running a censorship resistant platform. Our platform matters.com has a few characteristic that are important for this story. First, the user can publish anonymously or under a shortim. They don't need to connect their real world identity to their writing. Second, when someone posts an article, the content is automatically stored on IPFS. Even if an account is banned or server is taken down, the article can still be found through its IPFS hash. In practice, we found that sometimes the Google search engine even ranks the IPFS version article of an original article higher than our own domain. Third is we experimented with different ways to support creators economically. For a long time, we use likecoin. It's built on the Cosmos ecosystem to let readers reward writers. But now, the Litecoin is already disappeared. So we are experimenting with USDT tipping on optimisms trying to support anonymously tipping while keeping things usable. And then fourth is that we are exploring the ways to give users move sovereignty connecting maybe to the fiverse looking at protocols like lens and letting context flow across different platform instead of locking into a single database. It sounds beautiful. Privacy, anonymity, censorship resistant, greater incomes and interoperability. But the reality is much more complicated. The reality is four issues. The first issue is safety for the builders themselves. Some of our developers were based in inside authoritarian regime. Over the past years, they ran into direct and indirect security problems. The environment became so risky that invatally all of our team member had to fully withdraw from the authoritarian regime. That is already a very concrete cost. We lost talent. We lost contacts. We lost how we operate and just to keep ourselves safe. The second issue is what I call the honeypot paradox. If you build a censorship resistant platform, you naturally attract people who have sensitive things to say. Politic descent, whistleblowing, taboo topics. That is good. We are serving our own mission. But it also means you are concentrating risk. Now you have a visible place where many dangerous texts leave together. Even if the architecture is more secure, the attention itself is still really dangerous. Of course, matters.com is blocked by the great firewall in China that is expanded. But even with anonymity, there are still risks. People reveal themselves in their writing style in the details they mention. If the personal security awareness is weak, their words can still be used against them. There were cases where posts published on censorship reception platform were later used as evidence in court. So even if we as a platform protect the data, we cannot fully protect people from the consequences of what they write. The third issue is attacks and spam. Because we do not want to censor legitimate content, the platform also becomes attractive to people who want to abuse it like scammers, build farms, porn spammers, coordinated disinformation campaigns. At one point, we saw that between 60 to 80% of new content was lowquality or malicious randomly generated spam scam campaigns or explicit content. At the beginning we did manual moderation but the cost was impossible. So we started to trim models based on human labels to automatically defect detect and d rank or hide that contents. I bet the decentralized publishing platform such as mirror or paragraph have focused faces the same problems. This creates a painful contradiction. a censorship resistant platform that now must actively filter and remove a huge amount of materials and articles. You defend, we defend free expressions against political censorship, but we still need some kinds of moderation against abuse. Where is the line? Who decide what to ban is a really difficult problem for us. A very vivid example happening during the lunar new year this year. We suddenly saw a massive wave of almost identical scam content. It was so big and so coordinated that it triggered automated anti-fraud systems outside our platform by the democratic government. In Taiwan, which is a democracy, but also struggles with fraud. There are automatic boats that look for scam patterns. when they detect a certain threshold they can trigger mechanism to block or downgrade the domain at the ISP or platform level. Because of the huge volume of scam post, our domain was auto automatically treated as suspicious for several days. People inside Taiwan could barely access also matters. Not just China but also Taiwan. It was exactly during the holiday period when people responsible was offline both on our side and inside government. It took time and personal relationships to get the block lift lifted. For me, this looked like a new kind of attack pattern using the spam and fraud content to trigger automatic safety systems in the democracy in order to block a censorship resistant platform without anyone even making a consensious conscious decisions. So on one side you have authoritarian firewalls. On the other side you have democratic anti-fraud infrastructures. Both can be used intentionally or unintentionally to shut down vulnerable spaces. The fourth issues is a mismatch between what we build and how most people actually use it. If you look at our user base, maybe just 10% of the users write about sensitive civic, political or investigative topics. The other 90% writes about daily lives, books, music, personal reflections, fictions, lifestyles. Those posts are often beautiful and valuable, but they are not a high risk. This is this has a funny implication for most users. The censorship resistant architecture is not the main reason they join. They just want a good writing community. But the cost and risk of running the platform are driven by the small group of highly sensitive content and by the existence of powerful advisories. We are too small to survive on the classic web two model of mass advertising like Facebook or Instagram. Our infrastructure cost is relative high. And from a pure commercial point of view, you could say why bother with censorship resistant at all if most content is normal. This is the economic side of the honeypot paradox. If you stay small and principled, you struggle financially. If you scale in a man way, you dilute the mission and may still get attacked. So where do we go from here? I do not have a grand solution, but I can share some directions we are exploring shared by these contradictions. First, we need to stop thinking of the platform as the only unit. um a single big visible platform is an easy target politically, legally and technically. Instead, we should build components a censorship resistance storage, privacy preserving identity, safer found funding routes, reputation system and interoperability protocols that can be combined in different way. In other words, we should get rid of a big beautiful platform more in more invisible infrastructure that makes that many small communities can reuse. Second, we need to eat our own dog food. When some of our colleagues had to leave their authoritarian places and their home, their model land, it became obvious that tool builders themselves are at risk. Their situation is very different from developers who get in trouble in demat democratic countries for money laundering accusation around coin mixer. Here is the people are threatens because they support free expression in authoritarian context. If we can't design the tools and processes that keep our own team safe financially, legally and physically, how can we claim the protect to protect activists and journalists? For us, this means thinking seriously about privacy, preserving payroll, safer collaboration channels, obsacultimizing centralized points of failure in our own organization. Third, we have to redesign incentives. Right now, the incentives are misaligned. An entrepreneur who works on censorship resistant carries huge downside risk and a very limited upside. Many activists and journalists who would benefit from privacy tools do not have the times, skills and money to support those tools. Many donors and governments want impact but they also want low risk and perfect compliance. We need new mechanism for funding and coordinating this work. Maybe the first is the channel for donation and grants that protect both the giver and receivers and the ways to build strutonomous but credible reputation for contributors. the governance structures where people from high-risk regimes actually have have a say without exposing themsel and without self censorship and the last maybe is the accountability that doesn't depend on dopings. Some of this can use crypto, ZKP or verifiable credentials. But the key point is not the buzz word. The key point is to design for human reality. The fear, the limited time, the limited attention and the low technical literacy. And the most important of all the existing trust network without technologies. It's a super important things because most of the journalists they don't rely on the the technology. And the fourth is we must close the usability gap. Most people at risk are not crypto natives. Many barely have basic digital security training. So privacy enhancing technology that requires complex key management, multiple wallets, and a mental model of blockchain simply won't reach them in time. If we care about real users, we have to prioritize default safety, clear language and workflows that fit into existing habits. Sometimes this mean accepting more centralization at certain layers while decentralizing the pieces that really matters for censorship resistant. Finally, we should be honest about the market questions. Maybe censorship resistant infrastructure will never look like a normal thought market. Maybe it is closer to building fire exist or seat belts or public libraries. You will not ask what is the TAM for fire exist. You just accept that they are necessary in a world where fires happen. That doesn't mean we give up on sustainability or efficiency. It means we combine different models. Some public funding, some philanthropies, some isical commercial services on top and some community owned structures. And we measure success not only by the revenue but by how many people can speak, organize and survive under pressure. Because of this, some of us are now exploring the idea of creating a new foundation focused on privacy first censorship resistant infrastructure for at risk communities in East Asia and beyond. In the long term, um in the short term, it might look like a research or coordination hub. In the long term, we hope it can channel resources, test tools in real context and make sure that what people need to need these tools, they are not working into the trap. We want to identify which privacy tool are ready to use today, which one are still too risky and how to connect them into real workflow for journalists, organizers and exile communities. And we want this work to be as independent as possible from any single government or single corporation because democracy themselves can be fragile. To the end, yeah, to the end, I want to return to the original questions. Does censorship resistant have market demand? My current answer is it has survival demand. People will need it not because it's convenient, but because the alternative is silence, selfcensorship, and exile without a voice. Our mission, I think it's all of us today, is to make sure that when the moment comes, the tools are there, the incentives are there, and the people who build and use them can stay alive both online and offline. If you are interested in exploring this kind of idea with us whether as a researcher, developer, founders or even user I would be very happy to continue the conversation after the sessions. Thank you so much. [applause]