How expensive should it be to run a full node? Current low-mid end hardware costs about 0.001BTC, and storage is about 10MB/sat. If the blocks are fill for a year, that's a little more than 5ksat/yr. 100ksat startup cost and 5ksat/yr is really quite cheap.

With hardware costs going down and performance going up, I think a block size increase could be warranted. Carefully and cautiously, we could try 2MB, and then 4MB in another four years, maybe keeping trend to a maximum of 32MB as the adoption reaches the end of the S-curve.

A balance must be struck between throughput and security. The narrow optima is widening between these as of now, and I'd rather a world in which self custody and everyday people interacting with L1 remains a thing for my children.

nostr:nevent1qqsz8annrezs6zfzrt4yaml3y4475tsx6l8hakuj5s99dfcgzws688cpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhsygy6ftx7h9u9vh38fyxu5ewg860k2az74mqangzqtffdrxxpfzv38vpsgqqqqqqs2nxlx4

Reply to this note

Please Login to reply.

Discussion

Dude we hit 1 sat/vB in the mempool this week.

We good

For now, yes. I'm setting my sights not on this year but in five when adoption is up 10x and again after that. The chain doesn't need to contain all the world's transactions, but I don't want L1 to be only for the financial elite either.

2300tx/10min/8B ppl = 1tx/person per 66year

64000/10min/8B ppl = 1tx/person per 2.4year

One of two things has to happen for a self-custody future: greater block sizes scaling partially with adoption or an L2 that is trustless and doesn't require direct and proportional L1 interaction.

Layers IMO. Always there are tradeoffs.

I want to explore drive chains as a way to alleviate L1 congestion while maintaining some level of self-custody. It's seeming more and more like the only way. DC+LN could really widen the adoption potential.

Again, this is for the future, many years from now.

I think there ought to be more users of bitcoin that node operators. I think a single transaction ought to be cheaper than running a node. The upwards fee pressure of 8 billion people putting transactions on a 7tx/s network would make each cost possibly a years salary of an average person.

A metric we could use for fees is their proportion to the cost of owning and running a node for a year.

Trading "your everyday node operator" for "your everyday bitcoin user" is a profitable trade.

That's not to say that every transaction ought to be on-chain, but that users need to actually be able to set up their a LN channels in a decent timeframe.

nostr:nevent1qqsw0vsd8l46ewglpv7evs4vtcc87ju2772t7g8rrhsh43xky8xdymcpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhsygy6ftx7h9u9vh38fyxu5ewg860k2az74mqangzqtffdrxxpfzv38vpsgqqqqqqs98jtjt

I agree, this is the biggest source of real FUD for me. I don’t think bitcoins value proposition remains in tact if almost nobody on earth can do a transaction. Then it’s just gold all over again.

Exactly.

I'm not a big blocker, but nor am I exactly a small blocker either. In the far future (multiple decades, centuries), maybe we'll see giant blocks to maintain the L2s and L3s, but it's a far cry from what BCH will end up being.

Using bitcoin L1 has to be be more common than running an L1 full node.

nostr:nevent1qqsvsx9xapvgyp3qrjfudym7j80fdnc6nt9mseuz02k0huf57vxf23qpr4mhxue69uhkummnw3ezucnfw33k76twv4ezuum0vd5kzmp0qgsf5jkdawtc2e0zwjgdefjus05lv469atkpmxsyqkjj6xvvzjyezwcrqsqqqqqpxd3ayd