Stay up to date on all things crypto and blockchain

\> Well we definitely disagree on a few things but I do appreciate all the time you put into having a conversation. Awesome. You're welcome. \> I still don't agree with the idea that blocks always need to be full, or that the mempool should have multiple blocks worth of transactions pending, but that may not matter much. The point is: fees need to be paying for mining. Ignore the current subsidy: bitcoin needs to run for at least another 100 years, otherwise you should consider it already failed. Thus there needs to be a healthy supply/demand market for block space. Supply is basically fixed (although irregular) so demand fully determines the price. Whenever the mempool is empty that means that people can bid the absolute minimum fee resulting in low security: a miner can earn more money trying to re-org the last block that had some fees rather than mining a near empty block with no fees. Or they simply turn off their mining machines until enough fees aggregate in the mempool. All that is already pretty bad, but it also means that a real 51% attack would become significantly cheaper. So there must be enough backlog in the mempool to fill several blocks with enough fee paying transactions. \> I'm also not sure how a hard fork is technically hard. The basic change may be very simple, but getting consensus is hard and then coordinating and signaling for the switchover is extremely hard and basically guaranteed to result in at least some split where some stubborn people, as well as lazy people that didn't update their software, to stay on the old chain. I would be surprised if that wouldn't take multiple years of rolling out. And then there's a known wishlist of bugs and annoyances that can only be fixed through a hard fork, so it would be a shame to not also fix all of those in the same hard fork. \> it hurts the system to have transactions that get stuck for days Not really. It's just the way people need to understand the system works. Manage their priorities and expectations correctly: pay for what guarantee you want: 10 minutes, one hour, 6 hours, a day, a week, a month.... and while waiting, keep updating the transaction to do more batching (more efficient size, thus lower average fee per UTXO) and use RBF and CPFP to upgrade to a faster class. That will have to happen no matter what the block size ends up being (for reasonable block sizes). For extreme blocksizes the other problems just break bitcoin anyway, so those are irrelevant to even consider: too large means no fees and prohibitive cost to run a full node. \> the amount of space required will be dependant on the number/size of transactions stored That's a very old argument and easily debunked: infinite size means .000...001 fee, which means everyone can backup their porn collection onto the blockchain and there would still be no security. Blocks not only must always be full for bitcoin to work, they \_will\_ always be full because there's an infinite demand for high grade distributed backup capacity. The only question is find find the sweet spot where the aggregate fee is high enough while still allowing full nodes to stay decentralized. IOW: the fee in bitcoin is also an anti-spam mechanism: too cheap and it will just fill with spam. \> Don't misconstrue that as a recommendation to raise the block size this some ridiculous level Looks like your level of understanding is already far past the level that you would recommend that. :) But don't be mistaken: there are many people that have said that and a lot that still do. Gavin Andreesen was pushing for 8GB if I'm not mistaken. He went as far as claiming that he did a scientific research by testing it on his laptop and it worked, so it was fine. We're living to this day with idiots and scammers that still cite that bullshit (same as with his testimony that Craig Wright was Satoshi). This is why it's important to pinch new and upcoming bullshit in the butt as soon as possible. Every day disinformation lives it spreads wider and then takes 10x as many days to get rid of again. \> but I have a lot of concerns for bitcoin It's definitely good to have concerns. A lot about bitcoin is still not fixed or even unknown. It's still beta and will be for another decade or more. It can still fail completely. It can still go to 0. \> only way for mass adoption is by going off chain. That not only depends on how bitcoin develops and what the capacity ends up being (including efficiency improvements and block size increases) but also on what you definition of mass adoption is. On chain is guaranteed to never be used for daily cups of coffee. And why would it? Why would the whole world verify 7 billion cups of coffee transactions every day? Bitcoin will have to find some equilibrium between cups of coffee and major international settlements on the order of $1B transactions between central banks and multinational corporations. I hope it will be somewhere on the lower end of that scale, but there are just too many variables and that's too far in the future to worry too much about right now. Either way, other use cases can chose from a wide range of off chain solutions from LN to liquid to statechains to discreet log contracts to others and even to more trusted ones. All of which wouldn't be even possible without bitcoin existing. And all of which are in many respects much better than current banks and credit cards and money transmitters (but those too will probably find their place in this picture and still exist in some form, they themselves in turn settling on bitcoin or one of the off chain solutions). \> Inevitably off-chain solutions will be centralized in my opinion Many are already not centralized (LN). Either way: there are many shades of gray that can be a fine trade off depending on the use case. Many more shades of gray that didn't and can't even exist without bitcoin. So it's always a win, even if not completely 100%.