 Dennis asks, should more people run a full node in the light of the recent contentious hard fork and user-activated soft fork? You should run your own nodes, so that you can independently validate transactions. If you are not running a node and validating the rules, you are not part of consensus, you are not participating in consensus. If you have to ask someone else if a transaction happens, and every time you type in the address of a block explorer, and you say, hey, has my transaction confirmed, what you are doing is delegating your ability to independently validate that transaction, and to find out if it is confirmed according to consensus rules, you are asking someone else and delegating that authority to someone else. That is centralization. Who knows? What consensus rules are they running to decide that? Are they telling you that this was validated on the fork that you think you were running, or maybe they are following the wrong fork? You should run a full node in order to answer those questions yourself. The reality is that most people can't do that. Technically, it is not easy, but more people can. In fact, it is much better if you can run a full node that operates on an independently controlled and paid-for internet connection, that is not running on a virtual private server, on Amazon web services, etc. If you have an internet connection at your home and you have the desire to use this technology, run a node. You can run a node on not a very powerful computer. It doesn't take that much to run a node. Any desktop will do. You can limit the amount of disk space you use, the amount of bandwidth you use, by supplying appropriate parameters and configuration options to, say, Bitcoin Core software, or you can run a different type of software if you want. You can run a node with just a tiny bit of bandwidth, by not keeping a full copy of the blockchain, but pruning it. That will give you the ability to validate. It will not give the network the benefit of your bandwidth and storage to help bootstrap new nodes and support lightweight clients and things like that. If you have a connection that is not capped, or that is capped at reasonable levels, like 200 or 300 gigs a month, which is often the case with cable providers and broadband providers, then you can offer a service for the network, which is keep a full copy of the blockchain and help nodes bootstrap and lightweight clients run. That improves the resilience, robustness, and security of the entire network itself. If you can do it, do it. It is a great experience. You learn a lot about all kinds of different technologies. You learn a lot about software, how to compile software, how to run software. Even better, if you really want to get your hands into a nice, meaty project, go check out the Blockstream Satellite program. For about $150, you can buy a 40-centimeter KU-band satellite dish. It is basically the same dish you see on houses for direct TV or one of those services. It is basically a TV satellite dish. You can buy one of those, put the receiver on it, and connect that to a USB software-defined radio dongle, which you can also buy. They sell them as a kit. Nice weekend project. Get up on your roof, put in a satellite dish, point it in the right direction, connect it to a desktop, plug in a USB software-defined radio, and now you have a node that is receiving blocks and transactions from a satellite, and can compare those against blocks that are being received over the internet, and perhaps also serve internet clients from your satellite feed. That massively increases the resilience of the network. Yes, you can do a lot to help increase the resilience of the network. Why is a satellite connection so valuable? A quick follow-up to the previous conversation. The reason a satellite connection is so valuable is because, even with power outages or internet outages in a particular region, if you have nodes that are connected to satellites, they will remain connected. The satellite replays information from the last 24 hours, so you get 144 past blocks. Even if you lose power for half a day, you can catch up with your satellite connection when you reconnect your power. If you lose internet for days or weeks, you can continue to remain synchronized to the Bitcoin blockchain over the satellite connection. This means that it makes the system resilient to shutdowns, even deliberate shutdowns, port blocking, and other censorship activities on the internet. When you regain your internet connection, or perhaps over a Tor connection, VPN connection, or a different port, you can serve those same blocks and transactions to your neighbors who don't have a satellite dish. You not only are resilient to censorship, but you also undermine the censorship of others by giving them blocks and transactions through various types of other networks like Tor or VPN. Lazar asks, what would be the requirements for someone to host a lightning network payment channel? Will this proof-of-stake model make the rich richer? I think the requirements are fairly lightweight. There is an implementation of the lightning network daemon that uses a light client called Neutrino. You don't need to run a full blockchain node, a node that stores the full blockchain. You don't even need to run a fully validating node. Instead, you run this lightweight node that validates the block headers and also validates transactions. It is a much lighter weight node than a full node. That, in conjunction with the lightning network node, can run on a smartphone. In fact, one of the participants in lightning development, a guy by the name Jack Mallor, has recently demonstrated a real mainnet lightning network transaction running on a lightning node, managing its own payment channels and even routing on an Android smartphone. You can run that. The lightning network is also going to run in the form of wallets. A wallet that runs a lightning network node can open payment channels to other network nodes. It can also route payments between them. It can either use a connection to a fully validating node like Bitcoin Core, running somewhere on a server that you control, or run a lightweight node like Neutrino in order to support that. The requirements are pretty low. You can run a lightning network payment channel and route lightning network payments without a very big computer. In fact, arguably, with less resources than you need to run a full Bitcoin node. Will that model make the rich richer? The idea that if you have lots of payment channels and you fund those with very large amounts of money, then you can collect lots of the lightning network routing fees, which allows you to run more channels. That has been one of the persistent fears. I think there is a pressure in the opposite direction. You can route payments that are smaller with just a few payment channels that are not very big. In terms of funding, you could take maybe a tenth of a Bitcoin and run 10 or 15 channels to a bunch of nodes. You can start collecting some small fees and increase your holdings. You think, why not run 500 channels and fund them with 15 Bitcoin and then do the same? Well, now, because you have put 15 Bitcoin in a system that is controlling the keys, the risk of losing that is much higher. There is this opposite pressure against centralization, which is that as you centralize lightning nodes, you increase the risk of theft. Just for the same reason that exchanges are less secure than the wallet that is on your smartphone, is because an exchange is trying to protect assets worth millions of dollars under the control of just a few keys, whereas you are just running a single wallet and it is a lot harder to hack that.