I've pointed out before that the enthusiasm for micropayments is from people who want to collect them, not those who want to pay them. Unlike credit cards, which, when introduced, were easy to sell to buyers but hard to sell to merchants, since the merchant pays fees.
Cryptocurrencies are not a magic bullet here. The technology for preventing double-spending is incredibly inefficient and slow. Transaction costs are high. Right now, a Bitcoin transaction costs at least US$5. It's been as high as $20.
>but hard to sell to merchants, since the merchant pays fees.
There is a nice article which hopefully somebody can link that explains that visa was a success because merchants reduced their costs, too. They had to do informal credit accounting for all their clients who only paid for the purchased goods once a month. With visa, that accounting position, which could be a full-time employee, wasn't needed any more. Furthermore, they reduced their risk of clients defaulting on those micro loans.
The actual cost of a Bitcoin transaction (when not socialized via inflation, or block rewards in the parlance) is ~the electricity cost of 700kWh, or $50, plus the cost of developing, building, and maintaining the miners. Each transaction also generates 87g of e-waste so that's gotta be non-trivial.
I'd suspect each BTC payment is actually $60-70. When block rewards end, someone's gonna have to pay that.
Ignoring the fact that's going to occur in 2140, what you've stated is simply not true, users decide what fees to pay, no one else determines the price.
I send zero fee tx's regularly. If you are in no rush the cost ranges from tiny to non-existent sending money across the planet.
It's never going to be used for micro transactions without a second layer but the whole premise of this comment is out of whack with the reality.
As I stated, right now, the direct transaction fees are low because the total fees are socialized. Miners make that back in block reward. Recently, people refused to pay more for transactions and BTC price remained constant-ish after the halving so hash power went down.
The actual cost of a transaction can be represented as the following equilibrium:
(E * pkWh) = F + (R * pBTC)
- E is the average amount of power required to process one transaction, a function of hash rate (~700kWh)
- pkWh is the average price paid by miners for power (~0.08 $/kWh)
- F is the average fee paid for a transaction directly (~5)
- R is the average block reward per transaction (6.25/3500)
- pBTC is the market price of a Bitcoin (~9500)
If (F + (R * pBTC)) is ever less than (E * pkWh) as happened after the halving, something's gotta give. Either the fee has to go up, the price of a Bitcoin has to go up, or the hash rate will drop. The halving demonstrated (3).
Make no mistake, though, the cost of a bitcoin payment is ~$50, you just have to distribute it over the parameters.
You didn't think miners were given you "almost free" transactions for their health or something did you? Your very-low transaction fee is the tip of a giant cost-burg.
As models go this is ok to highlight the cost/reward incentive to mine at all, but with the original whitepaper we all already knew that full network hash rate depended on cost of electricity, the block reward, the fees of transactions in the block, and the purchasing power of bitcoin. What's new, other than an incorrect focus on "per transaction" when the real cost is system-wide and depends on different/more parameters (sum of F, difficulty rather than hash rate, and transactions per block which individually contribute arbitrary amounts including 0s to the sum of F)? It just seems like another tiresome way to make the years-old remark of "wow look how much electricity the full system (a decentralized trustless ledger) consumes and how few transactions per second it supports!"
> It just seems like another tiresome way to make the years-old remark of "wow look how much electricity the full system (a decentralized trustless ledger) consumes and how few transactions per second it supports!"
The only way it doesn't waste inordinate amounts of power and support a reasonable number of transactions is if you pretend.
It is an inordinate amount of power relatively speaking and the transaction throughput is sad (though not sad enough to get people to use other versions that make different tradeoffs, apart from the obvious one of mainstream finance with Visa, Paypal, et al.) So what?
Waste? It's evidently not waste, because miners are compensated. Missed opportunities for that energy to have done work on something else, or at least done more efficiently so the surplus could do something else? Perhaps, but that can be said of anything. I'd instead criticize the far greater missed opportunities of not coating deserts with solar panels, not peppering the world with nuclear reactors, not having satellites beaming microwaves down, etc. etc. We'll never become a Type 1 civilization with attitudes suggesting we lower global energy consumption or try to allocate it only in approved ways.
> It is an inordinate amount of power relatively speaking and the transaction throughput is sad (though not sad enough to get people to use other versions that make different tradeoffs, apart from the obvious one of mainstream finance with Visa, Paypal, et al.) So what?
Well, I mean, if you're trying to launder money or manage a ransomware product, PayPal and Visa aren't great candidates.
> Waste? It's evidently not waste, because miners are compensated.
People are compensated for all sorts of wasteful garbage. I could pay some dude to dump gasoline directly into the rain gutters. Or move a mountain 6 feet to the left using only a shovel. That doesn't instantly turn it into a productive activity.
> Missed opportunities for that energy to have done work on something else, or at least done more efficiently so the surplus could do something else? Perhaps, but that can be said of anything.
There's no massive energy surplus. If we turned off the 55TWh/yr that's being burned on multiplayer Excel, it would stop a non-trivial percentage of our greenhouse gas emissions.
> I'd instead criticize the far greater missed opportunities of not coating deserts with solar panels, not peppering the world with nuclear reactors, not having satellites beaming microwaves down, etc. etc.
Sure, I'm down. Sounds great!
> We'll never become a Type 1 civilization with attitudes suggesting we lower global energy consumption or try to allocate it only in approved ways.
Not with that attitude :)
To me, it's like there's a giant smoldering tire fire in the town square, and everyone's walking around praising the tire fire for the nice warm glow it creates.
How did you arrive at these numbers? By comparing # of transactions in a block, to the block reward + fees paid for that block? Or by comparing the amount of energy consumed by the network in total and the number of transactions processed by the network?
Cause fee replacing subsidy is the only way to ensure you can maintain the security without inflating the supply. Increasing the blocksize is not an opion. The solution is Lighting Network that allow cheap instant micropaymemts and leave large txs onchain where a tx is more expensive. LN is where most of the innovation is happening right now.
Which means they’ll invest more to produce it, providing greater security to the network by making it more expensive to conduct a successful double spend attack. The users also want it to be secure. It’s a great alignment of incentives.
Since it is a micropayment, it is possible (or acceptable) to accept payment with zero confirmation (in less than 10 seconds), as pointed out by Satoshi in the "Bitcoin snack machine" post:
https://bitcointalk.org/index.php?topic=423.msg3819#msg3819
Yes, there is a risk of double-spending, but since this is a small amount, the cost of the attack may be higher than the benefit. The payment processor can also deploy more full nodes in different geographic locations to monitor and detect potential double-spending transactions in the transaction memory pool.
Since all you need is the software to generate your fake transaction with the right technical format, the cost of the attack can be zero.
And then a million fraudulent microtransactions will add up. Of course, that would never happen in practice, because the Blockchain cannot handle a million transactions in any reasonable time frame.
In case there are multiple conflicting transactions (for double-spending purpose), the network will only accept the first one and reject the rest. So the micropayment processor needs to wait for a few seconds when it receives the first transaction in the mempool. If no other conflicting transactions arrive, the first transaction will be acccepted. These few seconds give the first transaction big advantage over the double-spending ones, as it will most likely be included in the block.
If the attacker sends the legit transaction and double-spending one at the same time, it will be detected due to the few seconds wait by the payment processor.
there's no protocol-level guarantee of that, both transactions are perfectly valid, signed transactions and it's purely dependent on the good behavior of node operators to reject these
and in fact it is quite common to accept "double-spending" since that is really the only way to increase fees to push through a "stuck" transaction. Otherwise you have to wait an indeterminate (potentially forever) period of time until it exits the mempool of all network participants
If there are two valid, signed transactions that spend the same unspent input transaction (UTXO), only one of them will be included in the block. This is indeed guaranteed by the protocol [1].
Yes but it is in the best interest of the node operators to behave well in that way, so it's reasonably trustworthy even without protocol-level guarantees.
Huh, I might misunderstand it, but that property appears to be broken in account-based cryptocurrencies like Ethereum. Since there's no concept of unspent outputs, what could protect ETH etc users in a similar way? Wallet transaction index?
There's a nonce on each account. If you see a transaction from the same account with the same nonce it's a double spend . Nonce can't be skipped either, so a double spend requires two transactions with the same nonce.
It is broken in Ethereum. In Ethereum, since consistency is enforced by transaction nonce rather than UTXOs, there is no way for a "watcher" to keep the payer honest by rebroadcasting a previously signed transaction output. Indeed, in Ethereum you can "cancel" a transaction in the mempool by signing a blank transaction with a higher fee and same nonce, which is by design.
> I've pointed out before that the enthusiasm for micropayments is from people who want to collect them, not those who want to pay them.
It might be just out of my ignorance for what the downsides are, but quite often I do wish I had a way to make micropayments, and I feel like lots of others do feel similarly.
Have you seen how much opening a lightning channel cost? And opening a second channel, if you cannot find a route to your merchant? How about refilling your channel? Or how to close it, if a malicious node wants to steal your funds?
That's right, it will all cost the same as a Bitcoin transaction.
And how about receiving a lightning payment? Well... The merchant need to have a balance so he can accept your payment(!) Yes, that's another Bitcoin transaction right there. Or if you're lucky you can find someone who'll give you liquidity (for a fee of course).
It's debatable if micropayments will ever take off, but lightning never will.
last I checked LN channels must be pre-loaded with funds, which would seem to kill the incentive to use them for spontaneous one-off BTC transactions between otherwise unaffiliated parties, which would be most of the transactions taking place. Who wants to put up $100 just to transact $10?
You don't need to have a direct LN connection with the content creator, just a path through intermediate nodes.
More likely users won't have to deal with all that because there will be micropayment wallet providers that can aggregate the transaction costs of buying bitcoin and maintain large value LN channels.
Sounds like a good opportunity for a hub to connect micropayments merchants and consumers. Consumers can open a $10 channel, merchants open larger channels based on volume, and the hub provides rebalancing services for a tiny fee.
well sure, why not. this decentralization thing is mostly a fiction anyway, might as well re-centralize our superficially p2p money network back into a p2h2p model, since that actually works.
The handy thing about LN is that we can have that model without having to trust the hub in the middle. If there's a disagreement of the state of the funds in the channel it falls back to regular bitcoin transactions to resolve it.
If such a hub does this too much then they'll have trouble getting others to commit funds to a channel to them.
The thing is -- and I realize the flaws in this argument before I even make it -- I would want to pay 5 cents for every article not to access it, but after I have read it. And it would have to be my decision: if I think it was a high quality article, well researched and reasoned about, perfect. If it turns out to be garbage though? I wouldn't want to pay.
I don't think postpaying is a necessary model, there could be a popup saying you don't have a subscription and to read the (full, probably) article is $0.05, would you like to pay this? I would almost certainly click "yes" on a lot of the stuff I don't wind up getting much value out of, and hey, 5c is naturally not that much so it balances out! And maybe that Isaac Chotiner or Bill Simmons (stay with me) 8,000 word'er will be $1. Package deals are easy. In fact, what with d-u-m-b cookie settings, email begs, and rest of the stack of "first 5 seconds" mandatory interactions, right now would be a great time to pop up a less-annoying payment dialog: "[Yes] [Account] [Sign Up] [No]"
I used to think that, then I realized I have also thrown change on the ground rather than carry it. If the article sucks that just hurts my chances of purchasing again from that author or publication.
> If the article sucks that just hurts my chances of purchasing again from that author or publication.
And this is why most publishers will use subscriptions for their normal content, except some dubious ones who will use clickbait headlines to attract one-time buyers.
> The technology for preventing double-spending is incredibly inefficient and slow. Transaction costs are high. Right now, a Bitcoin
This is true of bitcoin. There are better cryptocurrencies which have already solved these problems completely. Have a taste on https://nanospeed.live/ to do a fee-less live transaction in milliseconds.
India seems to get away with it - with UPI sending around Rs 1 (~ 1.5c) for 'verification' etc. is routine, and as with any UPI transaction it's all free.
The only thing lacking with UPI is proper integration with merchants; large retailers rely on weird locked in services like PhonePe etc. for integration with their billing systems, but the rest seem to use a QR code and manual verification of payment.
> The only thing lacking with UPI is proper integration with merchants; large retailers rely on weird locked in services like PhonePe etc. for integration with their billing systems
My local grocery store (a small mom and pop operation) uses a QR code system with a speaker the recites out loud any new transaction so that both the merchant and the customer can verify it easily.
And besides, integration with merchants is not the job of UPI. It's the job of the application implementing UPI.
Mandatory note that Bitcoin is not representative for cryptocurrencies, and transaction costs are only high because it's been artificially hindered.
0-conf is also reasonably safe for small amounts. We must always remember that credit cards are perfectly acceptable, despite them only being "confirmed" after several months (so they cannot be reversed any longer).
> I've pointed out before that the enthusiasm for micropayments is from people who want to collect them, not those who want to pay them.
Consider me one of the mythical latter people. I really really really want to support quality journalism, but I don't spend enough time on any one news site to justify a monthly subscription (most of which are fairly expensive). I would be totally fine spending sub-$1 per article read, though, if that were an option.
I understand why it is not an option, but that's not really the point. If it were, I'd do it. Hell, I'd pay 50c per article just to avoid the annoyance of finding a paywall workaround.
I frequently come across paywall'd articles I'd like to read. I'm not willing to add more subscriptions to my life, but it would be great if I could pay 25¢ just for access to the one article. If the payment process was easy, I'd use it a lot!
Who in their right mind would need micropayments like this?
Of course, any reasonable solution would take, say, $10 from you, and allow you to just press "pay" anywhere a paywall is encountered. Of course it would just auto-pay for articles on sites you put in a whitelist, within a limit you set. All the clearing is done in non-micro amounts, say, monthly.
Such solutions do exist. The problem, of course, is the buy-in from site owners. I suspect that running better ads just has a higher ROI.
lightning is really only cheap if you use a single centralized node as a payment provider, in which case you lose the benefits of decentralization. Otherwise you have to frequently open new channels as they become depleted of funds, which becomes quite expensive.
and if you just leave one big channel with lots of money in it, then you end up with this fun situation! Hope you don't have a power outage... ever.
Even if you use a centralized node, this kind of centralization in LN is not as bad as the typical "centralization problem" often referred to in cryptocurrency, because it doesn't mean counterparty risk. The worst thing that could happen is that you have to wait some time to get your money back.
You can replenish channels, even multiple all at once, either through someone just paying you, or an on-chain payment through a "loop". And trading some decentralization for "performance" when dealing with small frequent payments is still better than other choices.
Every crytocurrency works well except for bitcoin, because its throughput was crippled to try to sell a centralized second layer owned by the company that weaseled their way into controlling the github and subreddit.
I don't think any of them make micro payments easy right now, but throughput is not a real problem until transaction volume goes WAY up.
Expanding throughput comes with severe tradeoffs: it increases the resource requirements for storing full blockchain history, and makes provision of mining services inherently more centralized in other ways as well. All things considered, a "second layer" with weaker trust guarantees may well be a preferable solution.
At the current scales, that is like saying a 128x128 jpg has big implications over a 64x64 jpg. Anyone who can watch a 240p youtube video can sync with bitcoin's limited blockchain. Not only that but most people don't need to and don't want to.
What is bitcoin's throughput? 2MB every -10 minutes-? That's less than 3KB/s. Let's not do this whole dance where the real numbers never come up in all the fear mongering.
PlayStation games need to download more than that to install. Many steam games push updates bigger than that.
A single 10TB hard drive would take over -92 YEARS- to fill up at that rate. And again, most people don't ever need to sync with the chain, they let servers do it for them.
And? For the third time, most people never even think about syncing with the chain. Do you understand how cryptocurrencies work? Have you ever used one? What is your point?
I've pointed out before that the enthusiasm for micropayments is from people who want to collect them, not those who want to pay them. Unlike credit cards, which, when introduced, were easy to sell to buyers but hard to sell to merchants, since the merchant pays fees.
Cryptocurrencies are not a magic bullet here. The technology for preventing double-spending is incredibly inefficient and slow. Transaction costs are high. Right now, a Bitcoin transaction costs at least US$5. It's been as high as $20.