41
votes
McDonald's serves 62 million customers every day
That is 718 sales every second.
The Bitcoin network can process less than 5 transactions per second, while consuming almost as much energy as the entire nation of Argentina. It would take miners on the blockchain 144 days to process just 24 hours worth of sales from McDonald's.
The math only gets marginally better for Ethereum.
I feel very uncertain about how people can think that this is a viable replacement for conventional currencies and digital payment systems.
To play devils advocate (I'll have my thoughts on the devil's advocate argument below), what a crypto enthusiast would say is that something like Bitcoin isn't supposed to be the layer on which most people make transaction. It would be like ACH in the US - the real backbone of monetary transactions in the states, but an extremely old and kludgy system that very few people interact directly with. ACH transactions take 3 days, and have the fantastic API of uploading plain text files via FTP (no, seriously).
Rather, people send money on the layer of payment processors: a paypal or stripe. See, when you pay, say, a merchant that uses stripe to handle payments, you don't pay the merchant directly. Rather, you pay stripe, then stripe pays the merchant. And that's where the illusion of somewhat speedy transactions comes in: stripe will keep track of the inbound and outbound cash that should be happening in its big merchant accounts. Then it will front the money for the merchant - your "actual" cash is still slowly being ACH'd to Stripe's big merchant bank account.
The actual ACH transactions are then done in batch - the payment processors figures out how much money should be leaving, how much money will be coming in, and keeps track of that internally. They submit big fat multi million if not billion dollar ACH transactions, and as long as everything goes smoothly then it's all fungible and no one cares. The risk they take when things don't go right (transactions get declined, refunds, etc.) is part of the service they provide for the fee that they take on every transaction.
The future for Bitcoin or other cryptocurrencies as a currency will be something like that - where it acts as the ACH of the crypto world. Your money will appear as if it is being transacted in seconds or minutes via the illusion of batching and fungibility.
Now, a couple of problems with that. First, batching ACHs, while an impossible poor developer experience, is significantly better than Bitcoin is. Bitcoin blocks have a maximum size, and therefore a maximum number of transactions you can batch (a little less than 3000). Whereas ACH, especially at the scale of a paypal or stripe where you get special exceptions by the banks, you can practically batch an unlimited number and amount of transactions and it's free. Of course, it also takes a fraction of the energy.
Even something like Ethereum, which does not have a maximum block size, but scales the amount of gas required to process the block with the number of transactions doesn't really work that well with the idea.
It's hypothetically possible for a new blockchain that's better suited for this to arise: one that's proof of stake, and has much better support for batched transactions. But it doesn't really seem like there's that much real appetite for crypto as a currency: seems like the current trend is as a "value store". So I'm not sure if it would ever happen. An ever appreciating asset is at odds with a usable currency.
Secondly, it would kinda dampen the whole "not centralized thing" when the usage of the currency is heavily dependent on what would likely be a handful of payment processors. Not a monopoly, but probably an oligopoly.
Third, as the world would likely still mostly move in dollars and yen and rmb, you'd still have the FX issues.
The Bitcoin ledger started on January 3, 2009. 12 years ago. Bitcoin has had ages to figure its shit out - and we are talking tech years here, not politics years or Valve years or etc. All of its problems were identified years ago. There were "fork wars", or some such, several years ago about people trying to fix these problems, and yet here we are.
I do think it's philosophically interesting, but I don't think that the core tech/principles will be meaningfully adopted. There are vanishingly few cases where you need a distributed trustless ledger, where it's also a feature that there is no controlling agent who can fix things when people fuck up.
It's because there's a strong incentive among crypto holders to pump up the value. Over the years this has translated to, essentially, PR campaigns extolling the universe altering potential of crypto and blockchain. Unsavvy tech journalists used to pick up the hype and publish articles based on it all the time.
More recently the hype has been around what crypto actually is to most users: a mix of gambling and investment. That, at least, is more honest.
The opposition is just a natural response to the level of bullshit around crypto.
That and the ungodly carbon footprint which, if we're not pretending climate change isn't real, makes using any high impact crypto unethical.
Not to say that decentralized currencies don't have potential value. I think they do.
There are still only the few technical solutions in use in any of them. Proof-of-work fundamentally doesn't scale. Proof-of-stake is simply another form of centralization.
From "No, your cryptocurrency cannot work":
Well, sort of. It's more like COBOL isn't it? It's got its roots in, how much production COBOL still out there? The vast majority of cryptocurrencies are Proof-of-Work, so basically the equiv of a 1990's car. Proof-of-Stake has even more economic problems at hand than Proof-of-Work (giving the most to those whom already have the most....hmmmmmm).
In the end, a crypto-Tesla is probably just Facebook/Microsoft/Amazon/Google/Apple announcing a "crypto" that is basically just a regular database that they've all agreed to share hosting of. 99.9% of non-speculative use of crypto behaves that way anyhow, because almost no layperson wants to deal with logistics of a secure digital wallet.
The only remotely useful crypto I've seen come around is Filecoin, and that's because it's a digital currency backed by a tangible, useful good. It's still consuming energy, but it's not pure fiat the way other crypto is..if anything it's just leveraging the existing infrastructure upgrades people would be doing anyway.
Far more than most would probably assume.
Wanted urgently: People who know a half century-old computer language so states can process unemployment claims
Thanks for that clarification! That is definitely what I meant.
LOL. NP. And yeah, I figured that was what you meant, but people who aren't working as programmers/IT would probably assume there isn't much 40+ year old code still in production environments... so it was worth illustrating just how untrue that is. :P
I do not think the efficiency problem is solvable without eliminating the mining component. All current "successful" cryptos actively incentivize people to make the system as inefficient as possible.
Ethereum's planned move to proof-of-stake could work or it could fail. A lot of users are protesting it, because I think they don't actually care about the philosophy of crypto, they just want their computer to print free monopoly money. Proof-of-stake has another big problem as well, in that the rewards for processing transactions are only paid out to individuals that own a significant stake. It feels very designed to make sure "the rich get richer".
The barrier of entry will be very low for proof-of-stake (unlike mining, it will have no high barrier of entry of building or buying many miners, having access to cheap electricity, and being able to run electricity-hogging miners all the time), and the profits are split between all the people involved, so if it's ever more profitable than other general investment opportunities, then more people will easily join and push the profitability down. Even ignoring the electricity savings, it's a huge improvement over mining in being more open to everyone.
My whole problem with crypto is that it seems to be a solution in search of a problem. And even for the few unique properties that crypto does have, no one has ever demonstrated to me why those small benefits are worth the horrible environmental impact. Is worsening climate change really worth having decentralized and unstable techbro money? It just seems completely superfluous.
As someone interested in the crypto space and in the technology itself, not just watching graphs go up, I agree that maintaining and defending old Proof-of-work chains (such as Bitcoin) is just not worth it and people who try to argue otherwise seem really misguided. However I do believe that better solutions that use significantly less power are worth investing in and developing.
This answer on the cardano subreddit is one of the best examples of some of the positives a decentralised system could have on the world that go beyond the usual "its money but no banks".
Not pro or against, I just wanna know when Argentina became a unit of measurement.
Americans will use any other unit as long as it isn't metric.
Since Evita.
So if there was a project that could handle the volume would you be less against crypto?
This is a know problem in the crypto space and there are a verity of SOLutions. From the BTC centric lightning network to high Transaction Per Second block chains.
There are projects out there that are attacking all problems out there that this thread mentions. I only have a slightly deeper understanding than the casual observer. So there are better resources out there then me.
Probably not, I find the idea of a currency that is immune to government regulation a fundamentally bad idea for plenty of non-technical reasons. I think once you solve these particular issues, you eliminate all the appeal of crypto for 90% of users.
Can I ask what fundamental issues you see in government-independent currencies?
It is hard to control things like inflation or deflation when you have no power over the supply of currency. Many of the levers we have at our disposal to mitigate recessions and deal with other financial problems go away without public control.
Crypto is not immune to Gov regulation... there are tones of cases where there have been crack downs. Just because crypto anarcho capitalist think they can skirt the law doesn't make it a reality.
Also to be pedantic, Federal Reserve is technically Independent from US Government.
The Crypto use case to me, is to create a system(s) that does not require 15-25% of Global GDP to be caught up in sustaining the global monetary system. Those global resources are better spent in other areas.