Oddbean new post about login | logout Stupid question (especially to miners) but will/should the block size ever increase? Seems like it should be on a set schedule like the halving, considering Moore’s Law but way less frequent.
Very reasonable question that even Satoshi thought would work However we already know blockchains don't scale, doesn't matter if it's 4mb or 32mb or a GB Increasing block size is basically just kicking the can down the road, eventually we need layer 2 scaling solutions So keep the blocks small to make node running as cheap and simple as possible and focus all scaling ideation on L2 My two sats nostr:nevent1qqs84xx0edf0ay9273zds5vaez4udx439t3cs5p9v2sr3uw3tavqdfgpz3mhxue69uhhyetvv9ujuerpd46hxtnfdupzq70pfuwdzux7q8vmct7qt324aln7j2790403mp85k668490z2vn0qvzqqqqqqyx38v2t
before you decide if you need a new thing, decide first if you can push more into your existing thing it's really not rocket science in economics it's called "marginal utility" nostr:nevent1qvzqqqqqqypzpckv7l8jqspl8u4y54dn9rcduwlrs4v2040nxce0m2h0cunvrj8tqqsdnesfhsr3xd38djcdf24lw7ljsqhyjphcx0lwyq3zzjxcfrg9keg5puueu
How good will LN and other L2s be, and how will people own their bitcoin if onchain fees are 500 usd for eg.
Very likely will have to do a trust trade off and use federations instead of using on-chain if they aren't willing to pay for that degree of security
If most people won't be able to own bitcoin because of high fees and have to use custodial solutions, then what are we doing? Because if people can't own bitcoin, they also won't run nodes. This is highly centralizing. Things are not as black or white, we will need both approaches imo, scale L1 and L2.
I think considering even at 1 sat/vB most people still gravitate towards custodians and not node running, the proportion of people node running and self custodying will remain constant over time. But that's all that's needed to have a sound currency system, because most people (normies) actually want a custodian Speculation though, I see merit in your points 🫂
Yes, people that choose custodians are not a problem, the problem is people that want to self custody won't be able to, due to fees. If what you say is true, that most people will use custodial solutions by choice, then we might be fine with L1 as it is, but it might be also difficult to know if many use custodial bc they want to or bc onchain fees are too high. I think the market will decide, in the end we all want bitcoin to be decentralized and succeed 🫂.
They scale if you don't keep all transaction history back to inception. Anything longer than 30days isn't needed. Beef the size, shrink the time, scrap the unneeded excess.
How is your node supposed to find your transactions from last year? Or verify that someone else's old coins you received aren't double spent?
You're constrained to Bitcoin. I'm talking a fresh protocol.
Drive Chains are that new protocol. Bitcoin is too well established, and the non-Bitcoin crypto space is too filled with 💩🪙's to establish anything new and legitimate. It just can't happen.
New nodes still need to sync. Multi-TB per year will get quite unwieldy, especially as we run into physical limits to data transfer and processing speeds when we're decades into this. Besides, you need a way to keep track of UTXOs to verify txs. Imagine when we're into the multi-deca-billions of UTXOs to track and verify. No go.
No you don't need to track UTXO's because UTXO's are stupid. Them and blockchain are the constraint of the whole system. The distributed ledger is all that matters. You don't need to know "who sent what". You only need to know "who owns what". That's it. The ledger. The logic is simple: Has key? Has balance? Send approved.
What if we embed another DLT into Bitcoin as a L2? Such as a Merkel tree balance proof protocol like MINA? Bitcoin can host the top-level proof of balance, and wallets will hold branch proofs.
This is kind of the idea pushed even farther with MimbleWimble on things like Grin, Beam, and LitcoinMWEB
Only ~18500 nodes accessible worldwide. I run 3 of those, and 2 of them are pruned. 99% of people will never run their own node, which to me largely invalidates the argument against larger block sizes. I would rather have larger blocks and smaller fees. If I truly want to be my own bank, I would have no problem adding hard drives to a server.
It matters how fast your node communicates with all other nodes globally. Consensus on the network takes time. Not to mention the storage problem.
I don't think there is a storage problem, and once again, pruned nodes would be fine for most. I wonder how many people are actually running full nodes right now. The speed is dictated by your internet. I imagine most people are not hosting on their own infrastructure either, but rather somewhere in the cloud, which will be highly accessible in terms of speed.
Good point @bitbybit. Decentralization is the limiter of speed. But even Litecoin has 2.5 minute block time. Not any more or less usable than BTC. But it is faster. To me, usable has to be 4 sec or less. I'm testing and learning Nano now. It's always been of interest to me. Checks all the boxes, fixed supply (all mined up front), instant transactions. I don't understand its mining/consensus yet. Maybe a dud. We'll see. I think Bitcoin has a wakeup call. It can remain as "digital gold", slow, store of value. But that means people use something else for daily transactions. Lightning, or another L1. To me, the usability difference between Lightning vs another good L1 is minimal. Risks, exchange time, costs. In fact Lightning probably fares worse at these. So if a faster L1 protocol comes out with moderate security, capital will flow there for payments. But it may start accumulating there and avoiding going back to Bitcoin at all if/when that L1 protocol proves itself. Very likely scenario IMO.
Interestingly that's less than accessible Monero nodes. Bigger blocks: Run LTC or BCH nodes Private bigger blocks: Run XMR node
I suspect people running XMR nodes care more about privacy than the average Bitcoiner. Sad, but probably true. So many maxis have never even downloaded Bitcoin core.
Not to mention that when a miner broadcasts a block they have an advantage over all other miners if they find a hash with a high enough difficulty. They broadcast that new block and are already working on the next block before the rest of the world knows. That's why you see alot of pools win multiple blocks in a row.
Exactly @Sovereign Bear At the crux, you're arguing more volume, lower fees. To achieve the same income as less volume, higher fees. Higher throughput. That benefits the network and users. Efficiency.
Before I became a maxi, I really liked the idea of MINA protocol. You've got a Merkel tree of account balances, and the top-level zero-knowledge proof gets updated across the network. The knowledgeable proof for each account is then held by the wallets individually. There may be a way to build this into Bitcoin non-custodially, such as publishing the top-level zk-proof on-chain and a manner to move in, out, and across L2 instances via an LN-like protocol.
Then why not 100Kb blocks? 1 or 2 Mb way too big. 100Kb is even cheaper and simpler. Small blocks centralize who can afford to transact on chain. People not transacting on chain won't run nodes. This also centralizes Bitcoin. There is surely a conservative way to increase blocksize without going crazy. Consider lower end of global latency, bandwidth, and storage based on current tech and go from there. Even Lightning eventually fails to work without ~100x increase in blocksize.
Keep blocks small and activate OP_CTV.
If you 10x the blocksize, it's still awfully small. You need to economize another way. I think something like segwit is acceptable though, because it incentivized a behavior that aided scaling. (ignore ordinals)
You want us to ignore that it also incentivized bad behavior LIKE ordinals? lol They really should have just simply raised blocksize to 2-4Mb instead of dancing around the what was the end result anyway 10x is small versus the entire world being users. But we dont have to support the world *right now* (if ever). That is not reality. There are far fewer user using the network (not a hypothetical) 10x is huge improvement for actual users.
Well, yes, but if we run into blockspace limitations, increasing blockspace won't fix much. Ordinals came and went, I don't think it was segwit's fault. Spammers gonna spam, as long as they pay for it it's fine. And I'm not running no 10 terabyte node btw. But the point was that the base layer going from 5-7 to 50-70 tps is not overly relevant in the context of scaling new financial solutions.
It was definitely aided by Segwit. You can argue it was worth whatever advantage Segwit brought, but "spammers" were given a massive discount (tbh not even sure if we should call them spammers since they paid the fee which *is* the spam deterrent mechanism). Without Segwit it would've made it not possible, reduced it, and/or made it significantly shorter. Anyone could easily say "I'm not running no 700+ gigabyte node". Why not 100kb blocks then? 10kb? It would make nodes cheaper to run and more decentralize. 5-7 to 50-70 tps, 10x, is huge for current users for the forseeable future wdym?
Keeping blocksize small is essential to maintaining adequate fees to support miners. Eventually, very few will transact on the baselayer, mostly nation states and large corporations. A "war" was fought over blocksize. Check out https://www.amazon.com/Blocksize-War-controls-Bitcoins-protocol/dp/B08YQMC2WM
Hey! I just noted about this, too! Self-custody is really important. Whatever self-custody L2 we invent will be undone of it requires the user to have direct access to L1. 0.5Mtx/day is our issue. If drive chains, or another BIP that proposes a way to self custody without touching L1, are needed. nostr:nevent1qqsxp7xy0y77gcyr23ecm2558e95fm5hjvjug8yd97mvt8u45zn2hvcpr9mhxue69uhksmmyd33x7epwdehhxarjxyhxxmmd9upzpxj2eh4e0pt9uf6fph9xtjp7najhgh4wc8v6qsz62tge3s2gnyfmqvzqqqqqqy77rtm8