A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network. And a 0-conf tx offers as much security as existing payment networks (which are reversible). When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away.
A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks. With Bitcoin he only has to wait 1 block (~10 minutes), reducing pressure on his cash flow.
A merchant must wait typically 60 days for a CC transaction to be considered irreversible. With Bitcoin he only has to wait 6 blocks (~1 hour).
So, comparing apples to apples, Bitcoin is either as fast as, or MUCH faster than legacy payment networks.
What is less clear, however, is the capability of the traditional payment networks to improve. Before bitcoin et al, they might have had no reason to. I don't find it very likely that the 60 days transaction speed, for instance, is unfixable.
Some people just needs to get off their asses and dive deep into some horrific legacy code and/or processes and fix or rewrite them. If they are pushed hard enough, they'll do it.
A 0-confirmation Bitcoin transaction similarly needs only < 1 sec to be broadcasted to the network.
No, payment networks verify that the user can spend immediately. That is full confirmation in a few seconds. Bitcoin fails dismally here on very limited transaction volume by taking 10 minutes at best for something approaching confirmation and has no chance with the current architecture of POW of ever realistically replacing those transaction networks - both because of energy use and time.
A merchant needs to wait 1-2 business days before he can spend the money he received through these payments networks.
Yes, Bitcoin has the advantage here - you'll get your money faster, though there is no real technical barrier to existing networks paying out faster, the banks are just incompetent and greedy. This does depend how the payment is done - using FPS it's available instantly, using cc it's not (often more like 7 days, not 2).
A merchant must wait typically 60 days for a CC transaction to be considered irreversible.
That is a matter of policy, not tech. If Bitcoin were a serious transaction protocol, customers would demand the same features, or a guarantee in law that they could return shoddy goods - either way, the protocol is not the problem here.
When I sell computer gear on Craigslist in person and someone pays me in Bitcoin, I only wait ~1 second for the tx to reach my phone's wallet before letting the buyer walk away.
In that case you have no idea if that transaction is actually confirmed and not a double spend (typically Bitcoin requires at least minutes to confirm), that's not equivalent to them paying you by bank transfer in seconds. Comparing a zero confirmation transaction from bitcoin to a full confirmation transaction from a bank is disingenuous at best.
That's not even touching on all the other flaws of bitcoin as a method of payment - irreversibility (yes people really don't want that!), pseudo anonymity (just enough for criminals, not enough to stop states), complexity meaning average users hand their entire savings over to someone else, lack of insurance or regulation (yes people really do want that!), lack of backing as a currency making it extremely vulnerable should it become popular to being targeted and cornered by states or money markets, hidden centralisation through miner cartels running the transaction network and a small group of devs/companies producing the complex software required, energy use of POW which means it could never handle enough volume....
Using merkle trees to store transaction ledgers is really interesting, but with the current architecture of bitcoin is not a rival to current payment networks mostly because it appears to be designed for something else entirely (to replace fiat cash it seems, which is on the way out anyway).
No, payment networks verify that the user can spend immediately. That is full confirmation in a few seconds. Bitcoin fails dismally here...
To call a credit card transaction a full instant confirmation is disingenuous at best. If I were selling goods I'd prefer the 0-confirmation Bitcoin transaction to a credit card transaction. And just like how you might refuse a credit card payment from someone without an id, you might also refuse a 0-confirmation Bitcoin transaction with no or very low transaction fees.
In that case you have no idea if that transaction is actually confirmed and not a double spend (typically Bitcoin requires at least minutes to confirm), that's not equivalent to them paying you by bank transfer in seconds.
That's correct, but it's not a fair comparison for his example, as no one on Craigslist will send a bank transfer to you. Judging from context mrb was comparing it to a credit card transaction, which you can do with PayPal and their free credit card reader that goes into your audio jack. If some stranger on craigslist wanted to buy something from me with either a credit card or a 0-confirmation Bitcoin transaction, I'd chose the Bitcoin transaction. Perhaps you underestimate the difficulty in performing a double spend. Or underestimate the amount of credit card fraud that happens.
You also mention the energy cost as a concern. It was estimated that the energy cost of running the Bitcoin network was $400 million in 2016. Bank of America listed $60 billion dollars in just operating expenses for that same year (and this doesn't include investment losses). Sure there is a lot more going on at BAC, but even if only 10% of that is spent verifying transactions, it still cost more than all of Bitcoin. And that's just one bank.
Since this is just the POW algorithm, increasing the blocksize or using an off-chain solution to increase number of transactions won't cause a scaling in POW energy usage. The POW algorithm costs just as much energy to confirm 1 transactions as it would 1 million. All that energy isn't spent doing math to confirm the transactions, it's confirming the "signature" on the block which lists all the transactions.
complexity meaning average users hand their entire savings over to someone else
Ignoring the fact that you've certainly handed your entire savings over to someone else, I'd like to see the sources for this. Things are so simple now. It's gotten so easy to store Bitcoins on your phone or computer. You don't need to download the entire blockchain anymore, and you can backup your entire Bitcoin wallet with 12 words, which you can choose to memorize or hide somewhere.
... making it extremely vulnerable should it become popular to being targeted and cornered by states or money markets
I would argue it's already become popular. And I think Bitcoin being banned by state entities is probably it's biggest vulnerability at this point, but with so many businesses involved in it, I don't see that happening. In fact, things actually seem to be going the other direction. Bitcoin has officially been made a currency in countries such as Australia and Japan.
... hidden centralisation through miner cartels running the transaction network
If these miners were in America, with 8 to 10 different companies mining the majority of Bitcoin blocks, no one would be saying anything. But when you have the same thing in China, suddenly everyone is afraid of secret miner cartels.
... and a small group of devs/companies producing the complex software required
The core Bitcoin software is open source. It can be inspected for vulnerabilities at any time, and with so much money in Bitcoin, it's being looked at very closely. Being open source also means people can make their own versions at any time. I know of two popular projects hoping to grab the official Bitcoin name right now. And sure if such projects get popular enough, and far enough along, the community could break out into an all out war. Each side trying to find as many vulnerabilities in the others' software as possible. But whoever wins will be the most organized, and the ones who makes the best, most stable, and most reliable code.
... it appears to be designed for something else entirely (to replace fiat cash it seems, which is on the way out anyway).
I do agree that Bitcoin has many traits of cash, such as irreversibility and pseudo anonymity. But I don't think it's "just enough (anonymity) for criminals". Banks can trace serial numbers on bills, and Bitcoin can be traced from transaction to transaction. Both can be traced when spent and you go in person or have something shipped to you.
However, Bitcoin's similarities to cash don't leave it dead in the water. Imagine what people thought when the first printed bills started being used, "Oh too many problems here for this to ever be a thing. What's to stop them from printing double the bills? Coins have intrinsic value and when I receive one I have the value instantly."
The "solution" to the scaling problem spoken about in the article allows off-chain Bitcoin transactions (by fixing transaction malleability). This allows for layer 2 services, and these will have a lot more in common with credit cards. The big one you've probably heard of it the Lightning network, which should allow for millions of transactions per second with complete and full confirmations in < 1 sec, and yes, with no chance of a double spend. There's no reason someone couldn't offer a service that also plays mediator for shoddy goods. Bitcoin might seem rather limited now, but it's by no means hit a technological dead end.
To call a credit card transaction a full instant confirmation is disingenuous at best.
It is fundamentally different from a Bitcoin transaction - it checks the user has the funds in seconds. I assume you're referring to refund/chargeback policies, that is a policy, not a technical limitation. CCs will be replaced, but I suspect not by bitcoin.
I agree in theory for most of the technical points on speed there are possible mitigations, but do think the design simply goes in the wrong direction in ways which are fundamentally flawed (e.g. POW, anonymous transactions, tokens for coins). It has some of the worst attributes of cash IMO.
But whoever wins will be the most organized, and the ones who makes the best, most stable, and most reliable code.
I find it far more likely if Bitcoin actually becomes mainstream that it would suffer collapse or hostile takeover as there is no backing authority, and lots of interested parties who stand to make a lot of money from manipulating the price of both transactions and the currency. I just don't believe the meritocratic tech utopia you paint here is close to reality - why would the best win? The bitcoin space is full of frauds and cartels already, even when it is not in widespread use.
My main problem with Bitcoin however is that it's a solution to a problem we just don't have.
Using transaction ledgers without users holding private keys is a feature. Central authority and reversible transactions is a feature (even if the store doesn't allow reversing, an authority should be able to add balancing transactions to reverse). Trusted and public identity of partners is a feature. Confidence in a currency backed by tangible goods is a feature.
The tech is interesting and something tangential may well develop from it (like verified public contracts, or public bets, or notarised transactions or similar), but I don't really want pseudo-anonymous monetary transactions with anonymous partners, or to possess a hoard of tokens someone can steal, based on an extremely volatile currency vulnerable to cornering. What I do need is fast verified transactions with known accountable partners on a stable currency, with the possibility of disputing payments and reversing fraud, which is what we already have with FPS, debit cards, or credit cards.
It's a difference that doesn't matter to the payment recipient. The sender can steal funds with charge back fraud. Period. Therefore the CC tx is not "fully confirmed". This is the #1 reason merchants love Bitcoin. A great solution to the very real problem of CC fraud.
It is fundamentally different from a Bitcoin transaction - it checks the user has the funds in seconds. I assume you're referring to refund/chargeback policies, that is a policy, not a technical limitation.
I'm not referring to policies, we were comparing the initial < 1 second transaction. You keep bringing up "checking the funds in seconds" but checking the funds doesn't mean anything if they are just going to result in a chargeback. What I'm talking about here is the trust you give that < 1 second transaction. And I honestly believe a 0-confirmation Bitcoin transaction more strongly guarantees value than a credit card transaction, and has a greater likeliness of ending up in your possession. And this is without any technological improvements on the horizon.
but do think the design simply goes in the wrong direction in ways which are fundamentally flawed
You've shown you might not understand how POW works in previous replies by claiming the power used for the POW algorithms scale with the number of transactions. A block with only one transaction takes just as much work as a block full of transactions. It's possible you are also misunderstanding other components of the design. Anonymity is also a common misconception. If you want to exchange your Bitcoin for goods or services, the person on the other end of that exchange can follow that Bitcoin all the way back to the miner who first mined it. There's no such thing as a hidden transaction. Sure you could launder the money, but that happens all the time with any currency. Alphabay, a large darknet market was recently shut down by federal agents who traced Bitcoin transactions back to their owners. If all currencies had the open ledger of Bitcoin it would actually make it easier to track these things along with other important things such as government spending or the workings of a nonprofit.
I find it far more likely if Bitcoin actually becomes mainstream that it would suffer collapse or hostile takeover as there is no backing authority, and lots of interested parties who stand to make a lot of money from manipulating the price of both transactions and the currency.
I'd still argue that Bitcoin is quite popular now. Rothschild just invested in Bitcoin just yesterday, and not Bitcoin companies, actual BTC through GBTC. There's a LOT of money in Bitcoin now. Bitcoin's market capitalization is roughly 45 billion dollars. Surely that'd be worth a hostile takeover right now if it were possible. And without a doubt there are a lot of parties interested in doing so. But it hasn't happened. The mining "cartels" you fear so much have just bent to the will of the people - that's the news in this article. This change allows for off-chain scaling of Bitcoin; something that goes directly against the miners self interest as they stand to miss out on fees for transactions done off-chain. This is something they've been fighting for years. And yet, here we are. (Also, I still maintain that if these were simply 8 to 10 American companies instead of Chinese, this wouldn't be nearly the issue it is to people.)
Prices of everything are manipulated all the time, with or without (and by) banking authority. You don't even have to get into controversial stock price manipulation to see this. You can just look at the gold and silver markets which always seem to be in the news, but made big headlines just in June for large scale manipulation. And Bitcoin is well known for being volatile, and has had insane price swings in the past, but that didn't make it collapse either.
I just don't believe the meritocratic tech utopia you paint here is close to reality
That's why I only brought it up last in my original rebuttal. I feel I was able to counter much of what you had said with the technology as is.
why would the best win?
It's really amazing to see. But it's that what looks to have just happened. Someone will probably write a book on all of this in the future. For a short oversimplified summary, "Bitcoin Unlimited" was a different group of Bitcoin developers who hoped to become THE Bitcoin developers. They made a competing client with different approaches to scaling and took a different technological path. These approaches were much preferred by miners. But bugs in their client were exploited several times, taking out a huge numbers of their nodes at once several times. The type of exploits that would have crippled the network had everyone been running their nodes. In the end, the community and economic powers sided with a group calling themselves "Core", as their code had fewer bugs and was more well tested. Their supporters were better organized and through market forces alone were able to force the miners to adapt a scaling solution that went against their own self interest. I suppose only time will tell if this was the right path, but I believe Bitcoin to be better now than before all of this went down.
My main problem with Bitcoin however is that it's a solution to a problem we just don't have.
A lot of things seem this way at first. No one realized how significant the internet would become. If you had asked someone 20 years ago if they thought the United Nations would state that internet access is a human right, they would have laughed at you, but that's exactly what happened just last summer.
Using transaction ledgers without users holding private keys is a feature.
You don't need to hold a private key if you want to trust your bitcoins with a 3d party. There are legitimate companies right now that you can use to store and exchange your Bitcoins. There probably always will be. Gemini is completely licensed and insured. But you keep overlooking how simple using a private key has become. You also mention holding onto tokens several times, which is also a thing no one does manually. Only the oldest digital wallets generated tokens randomly, nearly all are "deterministic" now and generated on demand. From just 12 words you can regenerate your entire wallet and have all of it's history and tokens restored.
Central authority and reversible transactions is a feature
I think central authority's advantages are up for debate. This is something that's been with us so long it's become scary to imagine a currency without it. It may not be a conventional opinion, but there do seem to be advantages on both sides of this issues. And it's a big part of what initially drew people to Bitcoin when it was new. The subprime mortgage crisis had just happened and the bailouts were on their way and a lot of people felt betrayed. Things like that happen around the world all the time. Right now Venezuela’s annual inflation is over 750% due in part to corruption.
No central authority can inflate Bitcoin’s price or devalue it by manipulating its supply. With Bitcoin, there is no risk of money printing (by the government, or others). There's no single point of failure. If the US Government were to default again, or fail completely, it could have harmful effects on the currency.
even if the store doesn't allow reversing, an authority should be able to add balancing transactions to reverse
This is also an issue for cash, but hasn't been a problem. Once you've given it up, there's no way to reverse that transaction if the store doesn't allow it. Of course services were built on top of cash to allow just this. Why is it such a stretch to imagine similar services for Bitcoin?
Trusted and public identity of partners is a feature.
You can associate your public identity with Bitcoin addresses. If you want to take payments only by people with public trusted identities, you're free to refuse any exchange of goods or services.
This trust goes even further though. Most digital wallets let you "sign" a message as proof of who you are or proof of what you are saying. You're able to "notarize" something just by signing it digitally. This is why all those people who have come out of the woodwork claiming to be Satashi (the inventor of Bitcoin) have failed. They can't sign a simple message saying this is who I am.
Confidence in a currency backed by tangible goods is a feature.
I'm surprised this is so far down your list. I think this is one of the greatest arguments against Bitcoin. Economists say being able to pay taxes with a currency is ultimately what gives it value. Bitcoin falls back on an older model of value, scarcity.
Bitcoin is still very young, and there's nothing wrong with being cautious and watching from the sidelines. But I feel a lot of what you're saying comes from simple misunderstandings. And perhaps complexity is an issue, but people don't have to understand payment gateways or settlement layers to use a credit card.
Volatility is certainly an issue, but if Bitcoin continues being adopted its volatility will continue to decrease. There are a lot of problems that still need to be solved to bring Bitcoin into everyday life, but it does offers solutions to problems as well. No more out of control inflation due to supply manipulation. Corruption can be spotted so much easier (in government or business) with a public ledger. It's also both portable and borderless in a time when the world is becoming more connected.
That's one way of putting it. Another way to look at the blocksize issue is: should the supply of transaction space be limited by free market mechanisms alone, or by committee decision/voting/negotiations?
There is an argument that block propagation and orphan rates acts as natural barrier to the blocksize.
With a very large blocksize limit, miners who produce blocks which are very big, run the risk of a smaller block from a competing miner winning the race, as it is faster to validate and propagate over the network.
This limiting factor should mean the blocksize converges to a size that the network can comfortably handle.
This isn't fundamental. The large-block miner can pre-stage copies of the block (before the nonce has been found) at a few replicas. If they successfully find a nonce, they blast the nonce to all of their replicas, which immediately disseminate the complete block.
This is where the disagreement lies. Should the price of on chain transactions be reduced and remain low.