Your Protocol Can’t Be Censored. Your Users Can.

Transcript

Everyone. Um Account withheld, no service, not found, app not available. These are just some of the many pop-ups that come up when you try to access restricted apps or websites. The thing that they all have in common is centralized powers that can determine when you see them and how you see them. Hi, I'm Isha. I run social distribution for crypto companies. So, distribution and discovery are literally my job.

Getting what you've built in front of the people that it is for and making sure that they can find it. Which which means I spend a lot of time thinking about how easily discovery and distribution breaks. We're building towards a permissionless future and we've put enormous amount of work into censorship resistance at the protocol level. Keeping networks running, bringing transparency into operations, making information really really hard to erase. A permissionless future also needs permissionless distribution and discovery. Something can be impossible to delete but still really really easy to bury. If you're building for the underserved or the underbanked, what happens when they can't even see that your product exists?

Say your currency is losing value fast and stable coins dollar stable coins are how you protect what you have. Then, overnight a regulatory body no issues legal notices to social platforms that force them to censor posts about dollar stable coins in your region. Nobody gets banned, no platform dies, these posts simply stop showing up. So, if it's not happening at the protocol level, where are users actually getting censored? There are three layers between your protocol and the people using them. Discovery distribution and application.

Every single one of them has a lock and you're not the one holding the key. Let's start with discovery. X opened source its recommendation algorithm in January and yes, I read it. It's actually useful. You can see how the feed ranks posts and has a ton of great insights. But, reading the post doesn't actually let you choose what the live platform runs. It won't tell you why you're not shown something and it doesn't give you anywhere else to take your audience. Transparency is great, it's just not the same thing as credible neutrality. Then distribution. In 2020, India banned TikTok, one of the biggest distribution channels in the country gone almost overnight. And last month, ahead of a protest in Delhi, mobile internet was cut for 24 hours across 4 km in the middle of the city. If you were at that

protest, you had no idea you had no data, no WhatsApp, no maps, no phone calls, no way to post about what was actually happening. The obvious work- work-around here is peer-to-peer. Bitchat, for example, lets you message over Bluetooth with no internet at all. But, in October, Apple pulled it from the App Store in India after a government notice and the same thing happened to the Google Play Store. So, even a tool built to reduce your dependence on connectivity can still be switched off one layer up at the App Store. Sure, there are work-arounds, but they assume a level of pre-existing knowledge from your users, which only makes them harder to access and makes for pretty bad UX.

So, where does this go? Someone made a point to me that I keep coming back to, which is fragmentation using the example of DeFi. The money didn't all move from centralized venues into a single decentralized venue. It spread into a million little vaults, exchanges, pools, each with their own operators. Distribution and discovery, they say, will fragment in the same way to combat centralized censorship. However, what we lose with fragmented distribution is community. There's no one place where everyone gathers. Discovery starts depending on knowing the right person to get you into the right group chat. That's the opposite of permissionless discovery that crypto was built on and the cypherpunk future was wasn't supposed to need an invite.

So, how do we build for this future? Let's go back to the three layers between you and your user. Ask, who holds the key? And if they said no tomorrow, what happens to your users? Can someone discover you without an invite? Not just whether an algorithm is open-source, but can people choose how they're ranked and can they see when something's being withheld? Two, if your biggest channel disappeared the way that TikTok did in India, how much of your audience disappears with it? If the answer is most of it, that's your single point of censorship and your single point of failure.

And do you have anything that works when the data goes off for your users? If the app store pulls you out the same way that it pulled BitChat, can people still get your product? A web app, a front end anyone can use, a PWA, whatever or maybe you need to build that first. And finally, can your community move without falling apart? That means identity and connections that belong to your users, not to one server or to one group chat or to one centralized entity. I'm not saying that centralized platforms are inherently bad. If I did, I'd be out of a job. But if we're serious about censorship resistance, discovery and distribution have to be a part of it. So, if no breaks any of the questions I posed, your users can be censored even if your protocol cannot.

Thank you. >> Now, I'm curious. What have you seen in terms of best practices coming along to counter this already in the space? >> Mhm. >> Can you give an example or something that you like? >> Yeah. I think uh BitChat was a great example. I think um there's also this project called the Offline Protocol that is working to combat this exact thing. Um they're all building to solve, I would say, different seg- segments and different parts of this problem. Um what we're missing is like one solution that kind of covers everything. Um but I think that that solution can be found by combining the best what we have that's combating each of these discovery and distribution problems.

>> This is the mic. Hello. Hello. Okay. So, after everything that uh Farcaster has went through over these years of decentralization, multiple platforms, now network state, >> Yeah. >> where do you think Far- Farcaster lies in all of this? >> That's a great That's a great question. Um I personally thought Farcaster was a great example of how this can look like and what it can look like. I think they had the tech ready. I think the P2P part of it was the missing piece a little bit. But they had the tech ready. They had a lot of all the different clients that could rank your feed differently. I think they solved they found a way to solve a lot of these problems and to make them accessible in very different ways. I think the part where Farcaster

kind of suffered a little bit is discovery. There was still and this goes back to kind of the UX question also that I posed of like people needing to have previous knowledge in order to interact with what you're building. It became a little bit difficult to onboard users who weren't already kind of in the crypto space. And if you need really good distribution platforms for a lot of the applications that people at Devcon are working on that take the best of the Ethereum protocol and take the best of everything that's available, best of security, best of privacy, best of sense of anti-censorship. You need to make it available to people who don't need to understand any of those words.

And I think Farcaster kind of failed in that one regard.