I have a fairly reductionist view on this, one I think is helpful, so here goes.<p>First, I think of almost all large-scale Bitcoin arguments as religious. Especially if you're an outsider to Bitcoin, it will probably help you to think of them that way, too.<p>People come to a belief, and then back-fill plausible arguments all the time. When there are vigorous debates like this in Bitcoin land, we usually have some strongly held underlying beliefs coming to conflict.<p>I would say the holy war here revolves around a single axiom on each side: the "Small Block" camp fundamentally believes that there is a sort of inflation that happens when block sizes grow. Just like they believed that increased adoption would increase Bitcoin prices, they believe at some level that increased scarcity for block slots will increase transaction fees, and that the system relies on this promise to maintain its integrity.<p>The "Large Block" camp fundamentally believes that the value in the protocol comes from cheap, large-scale transaction support, and that artificially limiting right now is harmful.<p>You can reduce most of the arguments to comments on this. In particular, I think the comments about centralized mining concerns are almost totally specious. Bitcoin has long ago left behind small hobbyist miners with tiny data connections. Block size changes would not have helped them keep up -- there are fundamental aggregation economics in mining that have pushed them out. Any small miner that wants to mine now will do exactly the same thing they would do with 1 Gigabyte blocks: get their work from a pool, and spot verify they haven't been cheated.<p>Embedded in the world view of most "Large Block" miners if you follow out the logic is the belief that probably transaction fees won't pay for the entire security of the network. Ed Felten put out some calculations to this effect while he was at Princeton. They're reasonably compelling arguments, or at least worth thinking hard about. If you give up on the idea that tx fees have to pay for the entire security of the network, I think large blocks become a super obvious choice to enhance the current utility of the network.<p>If you do not want to give up on that idea, then any block size increase decreases the chance that the tx fee scheme will work. Since it is currently not working (the last two blocks as I write this had .09 and .02 BTC in transaction fees respectively), all such changes must be vigorously fought.<p>At any rate, this is why the "small block" folks have a 'purity' vibe when they argue; I think they feel they are protecting the original scheme, one they bought into and care about.<p>As you might be able to tell, I don't think the tx fee scheme is working right now and hence am a "large block" kind of guy. Forcing it to work, (e.g. shrinking block sizes dramatically, and perhaps refusing to mine on blocks with insufficient fees) would dramatically lower the usefulness of Bitcoin to many people. As it is, it's super annoying to have to wonder if a transaction will get included in the next block.<p>Finally, there are some possible tx fee solutions orthogonal to block size miners could implement easily. I have thought for a number of years that miners should just have 'express, regular, bulk' time/fee cutoffs for txs they would include. They should publish these rates. Of course, this could only be done by larger pools, or a consortium of smaller ones. Even 20% would make an impact on the fee schedule.<p>When I think about the economics though, I ask myself 'who would it help?' Miners are already dealing with very difficult economics at a macro and micro level, especially if they have chip projects. Getting a little more juice out of their mining would be nice, but is absolutely not what the large-scale mining game is about. On the flip side, it adds cognitive load and frustration to people transacting. It doesn't make a ton of sense to me.<p>If we're in a world where Bitcoin is still struggling to be useful to a billion people daily, then cognitive load and frustration should be ruthlessly excised from the technology until we get to critical mass.