ot run segwit. [Audience applause.] We will build up a non-segwit pool and we will fund it, and if we see segwit transactions, we'll reject them. That simple. So, I talked about a cartel. I'm not going to run the cartel, but we're going to reject segwit transactions. So if someone else runs it, I don't care. But I don't accept segwit; I'm not going to. And we're going build a pool that doesn't do that. And anyone who wants to join in that, great. If we see segwit, we'll reject it. Out of hand. Full stop. And we can do that---you advertise both parts of the chain, so we will watch for segwit transactions (not pay-to-script-hash generally---segwit) and anything segwit, bye-bye. Which means you're slower. Because we will take at least 20% of the market, and we will do that, and we will build that pool, and it's coming very soon, and if you want to be slower, run segwit. [Audience question, hard to hear, "Will that require a hard fork..."] I don't need to include it [segwit]. I don't to hard fork. I, as a miner, choose. People don't choose what I do as a miner. I can reject a block any time I want. It is about time that the miners started realizing their role in this industry. The miners set the rules. Not every user. Not everyone out there. Network distribution. You're all wrong. Giant node versus outlying wallet. These [points to slide] are connectivity graphs. What we have in Bitcoin is, as we get more and more connectivity, we get more and more reachability. This is how it actually works. What we have in Bitcoin is not all these other things where people are talking about "power-law", "scale free", et cetra. They are not Bitcoin because Bitcoin was always economic. Any node gets more connectivity over time. That is part of the protocol. The bigger your node is, the bigger the number of connection pools. If you run a Raspberry Pi, you are cutting yourself off from most of the network. In graph theory, it is not how many vertices, it is the edges that matter. It is the connectivity between the graph that matters. So, you're all getting it wrong. Very simple: you don't do node count. Node count is shit. It has zero relevance. Read the mathematics. Zero. That [points to slide labeled "semi-complete ring"] is Bitcoin. That is the actual model of Bitcoin. Those gray areas in the middle, that is the connectivity between the main nodes. It is that dense. There's a distance in bitcoin of d = 1.32. Later this year, I'm not sure how we're going to do it, but we're going to start releasing data. I'm not sure how because we have petabytes. [Points to another slide with same title] That is also part of it. It is bigger than a small world model. It is not anything that people are talking about. It is not a mesh. You send one transaction in 2.3 seconds, once it hits the network, 99.98% of the hash power has your transaction. It can take 25 seconds to get to 99% of the network, but it's the hash power that matters. Who gives a rat if your little Raspberry Pi gets it late? You're not going to put it in a block any way. If you don't have it in a block, it isn't part of Bitcoin. That's drilling into that network. This is what people think it is. This is what Lightning is. It's a mesh. Lots of little hops, central nodes, et cetra. Red balloons in Bitcoin: a group of Microsoft mathematicians went over and did the mathematics behind sybilling and other things. Any network with a distance of three plus can always be sybilled. Mathematically proven: always. Every time you do it. That---Lightning---can have eighty hops. Not eight, eighty. That means it is always vulnerable to attacks. Not sometimes. Always. It is mathematically proven. Read the paper; read their results. Bitcoin doesn't get sybills because it has a distance under three. What we end up, with this, Bitcoin... the problem, people don't get the difference in words. Bitcoin is "highly centralized" because it has "high centrality". That Lightning mesh is more of what we are arguing is centralized. In mathematical terms is more distributed than any network globally has ever been. It has