11 ms·

I refuse to watch this 2.5-hour video because I don't trust his words. I watched the entirety of one of his previous videos on a friend's recommendation - "Lin

by nayuki 3y ago

I refuse to watch this 2.5-hour video because I don't trust his words. I watched the entirety of one of his previous videos on a friend's recommendation - "Line Goes Up – The Problem With NFTs" https://www.youtube.com/watch?v=YQ_xWvX1n9g https://www.youtube.com/watch?v=YQ_xWvX1n9g . While I agree with the main premise that NFTs and cryptocurrencies are very problematic from many angles (financial, hype, scams, etc.), he made so many factual errors that I found it very hard to take it seriously. I have worked with Bitcoin software and can point out things he said that are an outright misrepresentation of the technology and mechanisms. (See also: https://en.wiktionary.org/wiki/Gell-Mann_Amnesia_effect https://en.wiktionary.org/wiki/Gell-Mann_Amnesia_effect )

malephex 3y ago
Please give some examples.
- jekrb 3y ago
  One I didn't see the other's mention is that in Line Goes Up, Dan calls secure-scuttlebutt (SSB), a p2p messaging protocol, a blockchain. IIRC Dan was relying on a sentence from an outdated, archived documentation site made by an open source contributor many years ago. The core SSB devs then had to try to convince people that Dan was wrong about their project. Dan refused to admit he was wrong, despite a lack of blockchain in the open source code, and just pasted a screenshot of the old docs.
- nayuki 3y ago
  To avoid the cost of sitting down for 3~4 hours to rewatch the full video while critiquing it and writing notes, I opted to skim the entire narration transcript and pick out some points. This is not a complete critique, but I'm bounded by how much time I can afford to spend on it. [0h01m15s] He talks about the 2008 Great Recession and frames it in terms of greedy banks offloading risky mortgages. But he fails to mention that the government entities FMAC and FMAE insured those subprime mortgages in order to encourage home ownership, which is what enabled this risk-taking in the first place. If the banks had their own money at risk rather than the government's money, they wouldn't have done it. [0h02m15s] > the biggest returns on a bond come from when it first hits the market, a new bond that creates new securities sales is worth more than an old bond that is slowly appreciating, but not seeing much trade. What? I'm pretty sure that's not how the bond market works. Bonds are priced fairly. Bonds being traded are sold at neither a premium nor discount to their fair market value. Also, the trading value of bonds doesn't appreciate or depreciate unless the interest rate changes. [0h14m15s] > How it works in Bitcoin, simply put, is that when a block of transactions are ready to be recorded to the ledger all of the mining nodes in the network compete with one another to solve a cryptographic math problem that’s based on the data inside the block. Effectively they’re competing to figure out the equation that yields a specific result Wording nitpick, but they the equation is easily known, but the values to solve that equation are what are hard to find. [0h14m42s] > Once the math problem has been solved the rest of the validation network can easily double-check the work, since the contents of the block can be fed into the proposed solution and it either spits out the valid answer or fails. If the equation works and the consensus of validators signs off on it the block is added to the bottom of the ledger and the miner who solved the problem first is rewarded with newly generated Bitcoin. Another wording nitpick. The verifier doesn't spit out the valid answer; it takes a proposed answer and spits out "valid" or "invalid". Secondly, the notion of the block being added to the bottom of the ledger glosses over the critical detail that every validator does so. Any validator can choose to reject the block, but it is not in their interest to do so because they would be going against consensus. [0h17m57s] > Because electrical consumption, electrical waste, is the value that underpins Bitcoin. Miners spend X dollars in electricity to mine a Bitcoin, they expect to be able to sell that coin for at least X plus profit. Wrong again, this alludes to the labor theory of value. It's the other way around: Because people are willing to put a value on Bitcoin for their transaction utility, miners are willing to spend electricity to prop up the system. [0h30m58s] > It’s an ecosystem that absolutely demolishes consumer protections and makes the re-implementation of them extremely difficult. He presumes that consumer protection is the only worthy goal. What about merchant protection? Search on Google for horror stories of eBay merchants getting scammed on PayPal by customers who lie about not receiving the product. Look at Louis Rossmann's rants about payment processors forcing merchants to eat the transaction fee even if a dispute is resolved in favor of the merchant. > the transaction fees on popular chains are so prohibitive that it’s pointless to use them on any transaction that isn’t hundreds, if not thousands of dollars The transaction fee on the Bitcoin network is definitely not hundreds or thousands of dollars. Look up any fee estimator and the exchange rate. https://ycharts.com/indicators/bitcoin_average_transaction_fee https://ycharts.com/indicators/bitcoin_average_transaction_f... [1h52m06s] > A deflationary economy punishes buying things, as anything that you buy today will inevitably be cheaper to buy in the future. This is factually true, and I'm pointing to the whole context without quoting several paragraphs. He shows the standard government economics bias that deflation is bad and inflation is good. Yet, he ignores the evidence. For example, the tech industry is massively deflationary. If you have $1000 today, you can buy a better laptop/iPhone/GPU next year if held onto this money for a bit longer. Yet, people still buy electronics. What gives? Well, there is a certain benefit to consuming in the present. It doesn't matter if you get x% more in the future, because someday you'll be dead. The point of saving money is to spend it someday, preferably before you're dead, unless you are generous about giving it to your kids. [1h53m29s] > We used to have a web where anyone could learn to write a webpage in HTML in an afternoon. You can still do that. Nothing obliges you to use 10 frameworks. We at HN can appreciate the basics, and there are tons of threads on this topic. Aside from the specific quotes, he alludes to capital and capitalism 10+ times. I know he has an anti-capitalist, probably socialist bent.
  - Jasper_ 3y ago
    A lot of this is ideological difference, which is to be expected, but let's take one concrete thing we can check here: > The transaction fee on the Bitcoin network is definitely not hundreds or thousands of dollars. Look up any fee estimator and the exchange rate. From the chart you posted, the current fee is roughly 2 BTC, which maps to $54,000. So that's thousands of dollars. When the video was published in January 21, 2022, the transaction fee was about the same, but 2 BTC cost about $70,000. So that's thousands of dollars. At its peak in April 2021, the transaction fee was ~60BTC, with a BTC selling for $60,000. So that's 60BTC * $60,000 = $3.6 million. So I guess Dan was wrong. Transaction fees on popular chains can range into the millions of dollars.
    - tracedddd 3y ago
      The Y axis is dollars, not BTC.
    - everfree 3y ago
      Bitcoin network fees have never been $54,000, let alone $3.6 million. The most they've ever been is two, maybe three figures. You must be reading the chart wrong.
  - UncleMeat 3y ago
    A huge portion of this is nits. A considerable amount of this is not challenging what is stated, but instead is just saying "well that's good, actually." > I know he has an anti-capitalist, probably socialist bent. Oh no.
  - Fraterkes 3y ago
    https://blockchair.com/bitcoin/transactions?s=fee_usd(desc)# https://blockchair.com/bitcoin/transactions?s=fee_usd(desc)# We don't have to eyeball a chart, the highest btc transaction-fees are public record and have (very rarely) ranged into the 10s of thousands of dollars. More significantly, you seem to have missed his point: he isn't saying that fees are hundreds of dollars, but that a 50 dollar fee is too expensive for any transaction that isn't at least hundreds of dollars.
everfree 3y ago
It's not just you. I'm not about to go rewatch that video to pick out the inaccuracies again because it's such a long video, but he definitely could have benefited from running his script by a software developer for fact-checking.
- parthianshotgun 3y ago
  Examples?
  - everfree 3y ago
    Like I said I'm not planning to go back and re-watch the entire video to make a list (I'd be writing an entire essay), but here's one illustrative example I noted the first time I watched it: "[Proof of Stake] suffers from delays when elected validators are offline and the system needs to draw again." It's easy to look at a poorly designed Proof Of Stake protocol and point out flaws. But since his beef is with PoS itself, I think it's fair to measure his claim up against Ethereum's beacon chain which is considered to be one of the better PoS designs: * Ethereum's validators are self-chosen, not elected. * There is a ~99.5% participation rate, meaning that 1 in 200 slots are missed. Offline validators are very, very rare. * When this 1 in 200 chance occurs and a slot is missed, Ethereum's elastic block size mechanism (EIP-1559) ensures that the chain catches up by effectively adding additional capacity to the subsequent blocks. * Whereas this slight unpredictability happens very rarely in Proof Of Stake, it happens constantly under Proof Of Work. This is because PoW is an inherently unpredictable poisson process. Thus, PoS is a substantial improvement over PoW in this way. * The result of this 1 in 200 fluke is that your transaction takes 18 seconds instead of 6 seconds to reach preliminary finality. And that's only for Layer 1 transactions which are by their nature high-value; Layer 2 transactions enjoy quicker and more predictable soft finality. I also remember there being issues in his economic analysis of PoS, an unwillingness to engage with scaling technologies (hand-waving it away as based on nothing), no mention of for example Secret Shared Validators or Rocket Pool as solutions to the solo staking minimum, no mention of the history or state expiry roadmaps when discussing the growing size of the chain. Those are just some more examples, reconstructed from some old notes of mine I found, without me spending too much more time on this.

by nayuki 3y ago

malephex 3y ago

Please give some examples.

jekrb 3y ago
One I didn't see the other's mention is that in Line Goes Up, Dan calls secure-scuttlebutt (SSB), a p2p messaging protocol, a blockchain. IIRC Dan was relying on a sentence from an outdated, archived documentation site made by an open source contributor many years ago. The core SSB devs then had to try to convince people that Dan was wrong about their project. Dan refused to admit he was wrong, despite a lack of blockchain in the open source code, and just pasted a screenshot of the old docs.
nayuki 3y ago
To avoid the cost of sitting down for 3~4 hours to rewatch the full video while critiquing it and writing notes, I opted to skim the entire narration transcript and pick out some points. This is not a complete critique, but I'm bounded by how much time I can afford to spend on it. [0h01m15s] He talks about the 2008 Great Recession and frames it in terms of greedy banks offloading risky mortgages. But he fails to mention that the government entities FMAC and FMAE insured those subprime mortgages in order to encourage home ownership, which is what enabled this risk-taking in the first place. If the banks had their own money at risk rather than the government's money, they wouldn't have done it. [0h02m15s] > the biggest returns on a bond come from when it first hits the market, a new bond that creates new securities sales is worth more than an old bond that is slowly appreciating, but not seeing much trade. What? I'm pretty sure that's not how the bond market works. Bonds are priced fairly. Bonds being traded are sold at neither a premium nor discount to their fair market value. Also, the trading value of bonds doesn't appreciate or depreciate unless the interest rate changes. [0h14m15s] > How it works in Bitcoin, simply put, is that when a block of transactions are ready to be recorded to the ledger all of the mining nodes in the network compete with one another to solve a cryptographic math problem that’s based on the data inside the block. Effectively they’re competing to figure out the equation that yields a specific result Wording nitpick, but they the equation is easily known, but the values to solve that equation are what are hard to find. [0h14m42s] > Once the math problem has been solved the rest of the validation network can easily double-check the work, since the contents of the block can be fed into the proposed solution and it either spits out the valid answer or fails. If the equation works and the consensus of validators signs off on it the block is added to the bottom of the ledger and the miner who solved the problem first is rewarded with newly generated Bitcoin. Another wording nitpick. The verifier doesn't spit out the valid answer; it takes a proposed answer and spits out "valid" or "invalid". Secondly, the notion of the block being added to the bottom of the ledger glosses over the critical detail that every validator does so. Any validator can choose to reject the block, but it is not in their interest to do so because they would be going against consensus. [0h17m57s] > Because electrical consumption, electrical waste, is the value that underpins Bitcoin. Miners spend X dollars in electricity to mine a Bitcoin, they expect to be able to sell that coin for at least X plus profit. Wrong again, this alludes to the labor theory of value. It's the other way around: Because people are willing to put a value on Bitcoin for their transaction utility, miners are willing to spend electricity to prop up the system. [0h30m58s] > It’s an ecosystem that absolutely demolishes consumer protections and makes the re-implementation of them extremely difficult. He presumes that consumer protection is the only worthy goal. What about merchant protection? Search on Google for horror stories of eBay merchants getting scammed on PayPal by customers who lie about not receiving the product. Look at Louis Rossmann's rants about payment processors forcing merchants to eat the transaction fee even if a dispute is resolved in favor of the merchant. > the transaction fees on popular chains are so prohibitive that it’s pointless to use them on any transaction that isn’t hundreds, if not thousands of dollars The transaction fee on the Bitcoin network is definitely not hundreds or thousands of dollars. Look up any fee estimator and the exchange rate. https://ycharts.com/indicators/bitcoin_average_transaction_fee https://ycharts.com/indicators/bitcoin_average_transaction_f... [1h52m06s] > A deflationary economy punishes buying things, as anything that you buy today will inevitably be cheaper to buy in the future. This is factually true, and I'm pointing to the whole context without quoting several paragraphs. He shows the standard government economics bias that deflation is bad and inflation is good. Yet, he ignores the evidence. For example, the tech industry is massively deflationary. If you have $1000 today, you can buy a better laptop/iPhone/GPU next year if held onto this money for a bit longer. Yet, people still buy electronics. What gives? Well, there is a certain benefit to consuming in the present. It doesn't matter if you get x% more in the future, because someday you'll be dead. The point of saving money is to spend it someday, preferably before you're dead, unless you are generous about giving it to your kids. [1h53m29s] > We used to have a web where anyone could learn to write a webpage in HTML in an afternoon. You can still do that. Nothing obliges you to use 10 frameworks. We at HN can appreciate the basics, and there are tons of threads on this topic. Aside from the specific quotes, he alludes to capital and capitalism 10+ times. I know he has an anti-capitalist, probably socialist bent.
- Jasper_ 3y ago
  A lot of this is ideological difference, which is to be expected, but let's take one concrete thing we can check here: > The transaction fee on the Bitcoin network is definitely not hundreds or thousands of dollars. Look up any fee estimator and the exchange rate. From the chart you posted, the current fee is roughly 2 BTC, which maps to $54,000. So that's thousands of dollars. When the video was published in January 21, 2022, the transaction fee was about the same, but 2 BTC cost about $70,000. So that's thousands of dollars. At its peak in April 2021, the transaction fee was ~60BTC, with a BTC selling for $60,000. So that's 60BTC * $60,000 = $3.6 million. So I guess Dan was wrong. Transaction fees on popular chains can range into the millions of dollars.
  - tracedddd 3y ago
    The Y axis is dollars, not BTC.
  - everfree 3y ago
    Bitcoin network fees have never been $54,000, let alone $3.6 million. The most they've ever been is two, maybe three figures. You must be reading the chart wrong.
- UncleMeat 3y ago
  A huge portion of this is nits. A considerable amount of this is not challenging what is stated, but instead is just saying "well that's good, actually." > I know he has an anti-capitalist, probably socialist bent. Oh no.
- Fraterkes 3y ago
  https://blockchair.com/bitcoin/transactions?s=fee_usd(desc)# https://blockchair.com/bitcoin/transactions?s=fee_usd(desc)# We don't have to eyeball a chart, the highest btc transaction-fees are public record and have (very rarely) ranged into the 10s of thousands of dollars. More significantly, you seem to have missed his point: he isn't saying that fees are hundreds of dollars, but that a 50 dollar fee is too expensive for any transaction that isn't at least hundreds of dollars.

everfree 3y ago

It's not just you. I'm not about to go rewatch that video to pick out the inaccuracies again because it's such a long video, but he definitely could have benefited from running his script by a software developer for fact-checking.

parthianshotgun 3y ago
Examples?
- everfree 3y ago
  Like I said I'm not planning to go back and re-watch the entire video to make a list (I'd be writing an entire essay), but here's one illustrative example I noted the first time I watched it: "[Proof of Stake] suffers from delays when elected validators are offline and the system needs to draw again." It's easy to look at a poorly designed Proof Of Stake protocol and point out flaws. But since his beef is with PoS itself, I think it's fair to measure his claim up against Ethereum's beacon chain which is considered to be one of the better PoS designs: * Ethereum's validators are self-chosen, not elected. * There is a ~99.5% participation rate, meaning that 1 in 200 slots are missed. Offline validators are very, very rare. * When this 1 in 200 chance occurs and a slot is missed, Ethereum's elastic block size mechanism (EIP-1559) ensures that the chain catches up by effectively adding additional capacity to the subsequent blocks. * Whereas this slight unpredictability happens very rarely in Proof Of Stake, it happens constantly under Proof Of Work. This is because PoW is an inherently unpredictable poisson process. Thus, PoS is a substantial improvement over PoW in this way. * The result of this 1 in 200 fluke is that your transaction takes 18 seconds instead of 6 seconds to reach preliminary finality. And that's only for Layer 1 transactions which are by their nature high-value; Layer 2 transactions enjoy quicker and more predictable soft finality. I also remember there being issues in his economic analysis of PoS, an unwillingness to engage with scaling technologies (hand-waving it away as based on nothing), no mention of for example Secret Shared Validators or Rocket Pool as solutions to the solo staking minimum, no mention of the history or state expiry roadmaps when discussing the growing size of the chain. Those are just some more examples, reconstructed from some old notes of mine I found, without me spending too much more time on this.