# Censorship-resistance ## Wall of shame 1 Obviously, which already exists, cluster 2 Release channels Play store and App store 3 Assume Ethereum is operational 4 Assume an internet connection 5 Whisper easy to DDoS (maybe not censorship) 6 Reliance on Infura quite a bit 7 Legal entity provides attack vector 8 We rely on a small number of (fiat) bank accounts to compensate core contributors 9 Organization design is somewhat centralized (not a DAO) 10 We don't have a strong open source community (dependence on CCs) 12 Non-standard ports make ISP blocking super easy 13 NAT traversal isn't implemented for Desktop to be bootnodes 14 GitHub as a single point of failure (node.js) .... ... ... ... ... - Requires Internet to operate - keeping failure local, right now more system on/off Post notes Oskar - intro - looking at how the principle came into existance Everyone-adding WoS thoughts Michael: let's start from single points of failure Ricardo: platforms and search engines - able to censore at ISP level, dictatorship. How do you get around something like that? That's what we are solving at a first level Corey: Ned and I talked about onramping people right from your phone - something needs to be done at connection level, P2P Ricardo: Mesh network Michael: Let's not get too deep into technical implementations Corey: the way we use the internet is too centralized, and that's prime terrain for censorship Ricardo: Bigger problem than the ISP Jacek: Focus on building something that would only fail locally - mesh is cool but use could still be penalized or blocked Michael: What can we do? Ricardo: Specific swarms working on it - ultra light client Michael: Move the conversation to App Store Jacek: Economic incentive vs enthusiasm specialized nodes that do these things incentivize people to run these services distribute through something like swarm users get access that way, doesn't matter where the service is distributed Ricardo: Apple doesn't allow downloads outside of their app store Igor: Different distributions we can try Corey: People who need it will find a way to have it - how much time should we spend building so many options? Oskar: How about the social aspect? Igor: At this moment we're mostly relying on CC's. It's hard for others to work/build on Status Ned: Go and see what others need - what blocks, where to start. Find the bone structure of all these elements, in these places where tech and language are different from what we use. Enable them to use their framework - rethink/reshape these elements Jazek: Contributors come when something is useful to them, hard to come into the project unless you're an ETH enthusiast. Hester: Some dev switched from looking at Status to develop for Status. Can't recall the exact details but they had more documentation available. Perhaps we need to focus more on that - we have lots more available now, and Studio can also help Igor: We have ENS documentation now, we should be pushing more blog content Barry: in voting we filtered out poles that were created twice - so every pole, when you create a new one is just sequential Corey: We can't give up user data if we don't have the data. If we work to NOT have any power whatsoever, we can't be forced into anything Michael: How would people react if at some point we can't have a legal company and can only be paid in crypto? Corey: I rely on the legal company to live a normal life Ricardo: We shouldn't even have one :) Jacek: Hmm classic compromise - same as Slack vs Status, you need to compromise in the sake of efficiency, up to a certain point Michael: Can we be honest with ourselves - maybe we need these legal companies and can't exist otherwise Ricardo: Maybe we should decentralize that too, and have a lot of legal companies Barry: Is the DAO an extention of our legal entity? Oskar: Let's table that til tomorrow - we'll have a session on that Michael: What are we gonna do if someone turns off our cluster? Oskar: we could (very hackey) run stuff locally, Adam and others are relatively close to have a contingency plan if need Corey: As soon as we have the option, we should kill the cluster, the same way we're doing with using Status vs Slack. Even if it's cranky at the beginning it will force to improve it faster.