5 ms·

AI is not in the same category in my book. Both may be overhyped in some ways, but blockchain technology was fundamentally a bad idea with no legitimate use cas

AI is not in the same category in my book. Both may be overhyped in some ways, but blockchain technology was fundamentally a bad idea with no legitimate use cases.

bottlepalm 3y ago
Bitcoin has proven to be an inflation resistant store of value enabling private and fast transactions to anywhere in the world - seems pretty successful to me despite being constantly derided by hacker news for over 10 years now.
- angry_albatross 3y ago
  The value will deflate at some point, it's kind of a mathematical certainty. We cannot extract more money from the system than we put in, since Bitcoin is a non-productive asset. It does not generate revenue the way that a business/company does. And someone has to pay for the electricity of the miners, so money is draining out of the system, making it a negative sum game that people are playing. I would also disagree that the transactions are fast or private.
  - bottlepalm 3y ago
    Private as in independent of going through any bank or institution like you would have to do to move money internationally in any other form. A store of value like gold, the dollar, or Bitcoin has nothing to do with generating revenue - only the intrinsic value of how hard is it to create more - dollars can be printed, a gold vein can be discovered tomorrow that would crash the price, Bitcoin on the other hand there's no chance of creating more than what's planned - which makes it a great store of value in turn valuable.
    - angry_albatross 3y ago
      One nice thing about gold is that we don't have to keep paying money for gold to continue to exist. It will sit on a shelf and not charge us any money for sitting there. Bitcoin exists on a network of computers that need electricity to run. If we stopped paying for this network to run, the bitcoin would stop existing, since the ownership of bitcoin really is solely determined by the ownership records on this computer network. That is my argument, that Bitcoin cannot maintain its value long term because the value is leaking out of the system in the form of electricity bills. Furthermore, we cannot let these bills become too small, or else the network becomes vulnerable to a 51% attack, so as a society we collectively must pay a large amount of "rent" on this store of value, which causes the value in this value pool to slowly deflate over time. We have been overcoming this drain so far by "investors" continuing to pour money into the system, but those investors cannot possibly get all of their money back, because it's been spent on electricity.
      - bottlepalm 3y ago
        By your own argument gold is expensive - expensive to secure, expensive to transport, slow to transport (time is money), expensive to verify. So yes it does cost money to hold and transact gold. So your same 'leaky' argument applies. Maintaining any store of value or currency is never free, but the benefits far outweigh the costs as now we have an inflation resistant medium on which to trade goods and services with.
        angry_albatross 3y ago
        Well I'm not trying to argue for gold being a good store of value either, it might not be. But another reason why I would prefer to have gold is that people make use of gold to make jewelry and electronics, so that creates a demand that helps to prop up its value. The point I was making is that in a passive state, just sitting on a shelf, I believe that the cost of maintaining ownership of all the gold in the world is orders of magnitude lower than the cost of maintaining the entire bitcoin network, and I think that must continue to be the case due to the threat of 51% attacks. (If the cost of running the bitcoin network drops below a certain threshold, it becomes profitable for a rogue actor to rent a large amount of compute power and force in some fraudulent transactions.) So I believe that the "leaky bucket" effect is stronger with bitcoin than it is with gold, and there isn't a similar real world use case of bitcoin similar to the manufacturing of jewelry like there is for gold to counteract this leak. Therefore, the total value held by all of the holders of bitcoin must be declining due to this leak, which counteracts the idea that it is inflation resistant in the long term.
        bottlepalm 3y ago
        Your passive analogy doesn't work because Gold needs to be secured and defended, which requires energy or other people are going to steal it. Just like Bitcoin needs to be secured and defended with compute for at least new transactions. A 51% attack will allow you block new transactions or double spend coins you have, not spend or steal other people's coins on the ledger because you don't have the private keys for those.
- wpietri 3y ago
  Almost every word of that is untrue in practice. But let's focus on the core part, "successful". Let's compare with M-PESA, a money transfer solution that started around the same time. M-PESA has steadily grown, doing 26 billion transactions last year. [1] Bitcoin was somewhere around 100m transactions for the same period. [2] That's about 0.5% of the volume. And its worse than the raw numbers suggest, in that the M-PESA transactions were things with positive economic impact, whereas a lot of the Bitcoin activity was driven by speculation or crime. And that's wildly smaller than the number of credit and debit transactions that happened over the same time, of course. Bitcoin was hoping to be "electronic cash", but the shift away from physical cash was toward the already existing digital payment mechanisms, not Bitcoin. Merchant adoption, always small, went into decline years ago. So no, Bitcoin was not successful in the sort of terms that match its initial goals or what was hyped in the early years. [1] https://www.statista.com/statistics/1139181/m-pesa-transaction-volume/ https://www.statista.com/statistics/1139181/m-pesa-transacti... [2] based on eyeballing this: https://ycharts.com/indicators/bitcoin_transactions_per_day https://ycharts.com/indicators/bitcoin_transactions_per_day
  - bottlepalm 3y ago
    Read my post, I said successful as a 'store of value' not 'electronic cash'. Bitcoin is more akin to gold - an inflation resistant store of value. There is a premium for transacting it which means you should convert it into a more inflation prone transactional currency if you want to spend it.
    - wpietri 3y ago
      Well look at those goalposts move. Bitcoin's original purpose was precisely electronic cash: https://bitcoin.org/bitcoin.pdf https://bitcoin.org/bitcoin.pdf But if you'd like to shift focus to why "store of value" is also wrong, I'm glad to. A store of value needs to be more stable than the thing you're moving out of. It also needs to be relatively liquid, and should have low transaction costs. But Bitcoin is very volatile compared with major fiat currencies, and also when compared with gold. Transaction costs are relatively high. The market is thinly traded, and is widely believed to be manipulated. Gold, in contrast, is more liquid, cheaper to trade in, and much better regulated. So no, Bitcoin doesn't make for a good store of value. People wanting to store value would generally be much better off buying index funds, which are also pretty inflation resistant, and also have a positive return. But if they want to avoid equity exposure, then the gold standard for this is, well, gold. Bitcoin is terrible by comparison.
sublinear 3y ago
It can be argued that probabilistic generative AI is about as equally worthless from first principles.
- angry_albatross 3y ago
  You are claiming that there is some sort of mathematical proof that this AI has no "worth"? There are a lot of people who are finding value from probabilistic generative AI, to the extent that they are willing to pay OpenAI $20/month to get access to the best model available, so I think that counteracts whatever sort of mathematical proof you might make. It's true that crypto has a lot of users too, but I think the difference is that these users are mostly being deceived into thinking that their involvement in crypto will make them rich in the future, when it will not. AI is giving people immediate value instead of making promises of future wealth.
- CSMastermind 3y ago
  That seems untrue to me? There are several very obvious uses for ChatGPT. Almost everyone I've seen use it can think of something they want to use it for. I've never had someone give me a plausible usecase for blockchains or crypto currencies. Once you push on the usecases I have heard it quickly devolves into "you just don't get it" or some kind of appeal to authority like "well all the VCs wouldn't be investing in it if it wasn't valuable, clearly they're smarter than you."
kayodelycaon 3y ago
The one legitimate use case of using blockchain for decentralized digital currency ends up being used by everyone who can’t transfer funds using normal routes. So most of the people using it are either bad actors or speculators Kind of like Tor. Bad actors ruin all kinds of things. People who only see technology will only go on to repeat the mistakes of the past because they lack the necessary domain knowledge to know better.
- logicchains 3y ago
  >So most of the people using it are either bad actors or speculators It's a matter of perspective. For many people around the world, the US government is the bad actor, the bad actor preventing them transferring funds through "normal" (US controlled) routes.
  - kayodelycaon 3y ago
    Yup. The only reason blockchain has any legitimate use at all is because governments are bad actors. Doesn't change that many of the people using the system are also bad actors of one form or another.
- nkuttler 3y ago
  > So most of the people using it are either bad actors or speculators "bad actors" according to their government, which could mean people who simply want the right to vote, or to live free from oppression, secret police, etc.
  - kayodelycaon 3y ago
    Indeed, what about all of the decent and moral people using digital currency for such reasons? Across all of them? A very low percentage. Fraud and speculation is left, right, and center. Unfortunately, only a few, like Bitcoin, can justify their existence on moral grounds. Hopefully the concept doesn’t get banned completely.
  - acdha 3y ago
    Even if we take this at face value and assume these are all dissidents seeking freedom, cryptocurrencies would be a bad idea because they force you to leave a paper trail for prosecutors and deal with intermediaries who could be suborned. If Iranian dissidents buy something with cash, they might find that the person they thought was trustworthy is secretly working for the police and they’re going to have a bad time but they would at least only have proof of that single transaction. If they use cryptocurrency, the police get likely years worth of transaction history and a list of everyone you’ve worked with.
sonicshadow 3y ago
I wrong
- Nextgrid 3y ago
  > Plenty of great ideas and easy-to-see use cases from that first paragraph alone. Such as? Blockchains enable trust-less and decentralized ledgers at the expense of major trade-offs (user experience, lack of transaction reversibility, proof-of-work to secure the network, etc). Most useful business happens off-chain in the real-world, so you need to bridge between blockchain state and real-world state using a trusted party which throws away all the decentralization and trustlessness advantages, so you may as well just let the trusted party run a conventional database directly, and avoid the major trade-offs. I can't think of many useful & valuable use-cases that happen fully on-chain with no off-chain interactions, aside from cryptocurrencies. As soon as you have off-chain interactions (which is most of blockchain usage outside of cryptocurrency applications), the value proposition of using a blockchain goes away and a database makes more sense.
rchaud 3y ago
AI is exactly the same to marks that don't know any better and think complex new technology = $$$. AI having use cases for megacorps doesn't mean that scammers won't leverage the hype to sell get rich quick scams.
sleepybrett 3y ago
Sure but that doesn't stop the managerial and engineering tiers of those businesses to jump ship to the new hype technology. The managers didn't know shit but sales in the first place, sell coins or ai no difference. The engineers, just another new tech to learn.
tornato7 3y ago
For better or worse, Crypto is legitimately the easiest way to create a financial application, derivative, exchange etc... any programmer can create interesting, useful, and novel financial instruments like Squeeth, crvUSD, PoolTogether, etc in short order. Good luck recreating those in TradFi in under a decade, let along making them interact with each other in atomic transactions. Of course, some people see this as a bad thing because it enables scammers to create all sorts of new and improved ponzi schemes. Others think it's a good thing because it speeds up financial innovation and levels the playing field between big banks and small startups. Some might argue that crypto is only useful for building apps insofar as it avoids regulation, but you also can't convince me that Wells Fargo is the future of finance. Banks could never create a financial playground that works as well as Ethereum, even using their centralized database. IMO, so many of the projects that have come out of the crypto space are awesome and promising, but investment in the space grew faster than projects could mature. Unfortunately that leads to users losing $100M when Joe Schmo's cross-chain bridge gets hacked, when it should have never had that much TVL. Just my 2c.
- arlcode 3y ago
  I'd argue that the difficulty in banking is not to get the technology working. Financial software is regular software with additional audits and checks to make sure it's safe against the flood of attacks it will receive. With the kind of money we are talking about, you can hire people with the expertise to reasonably protect you against bugs and software exploits (something most traditional financial companies have a close to 100% track record in but many crypto projects failed). But after that you have a mountain of issues to consider that have little to nothing to do with software: - financial logical holes like flash loans being used to extract money "democratically" - people committing age old scams "but on the block chain" - founders not understanding problems such as that you cannot secure one unsecured coin with another one no matter the algorithm. - all sorts of unpleasant people using your "financial playground" to do things society frowns upon. All of these are solvable but not by choosing a better technology stack. Look at how much traditional institutions are spending on compliance and realise that you probably won't be able to cut that by an order of magnitude. Crypto is learning very quickly that most regulations do not exist to "keep the little man down" but because having regular people get fleeced over and over by charismatics liers/fools can have a devastating effect on any community.
RestlessMind 3y ago
> but blockchain technology was fundamentally a bad idea with no legitimate use cases. Or, you live in a bubble and haven't seen any useful examples. I have compiled some for you: https://news.ycombinator.com/item?id=32406095 https://news.ycombinator.com/item?id=32406095
- angry_albatross 3y ago
  These articles are mostly promoting bitcoin as a viable currency for people in desperate situations, and yes, possibly, if you are in an extreme situation like being extorted by someone with ransomware, or in a place where your government is falling apart or targeting you as a criminal, you might have some reason to use cryptocurrency in this way. But since it does such a terrible job as a currency (extremely inefficient, slow, and irreversible transactions, price volatility, no intrinsic value, etc) it should really always be a choice of last resort.

AI is not in the same category in my book. Both may be overhyped in some ways, but blockchain technology was fundamentally a bad idea with no legitimate use cases.

bottlepalm 3y ago

Bitcoin has proven to be an inflation resistant store of value enabling private and fast transactions to anywhere in the world - seems pretty successful to me despite being constantly derided by hacker news for over 10 years now.

angry_albatross 3y ago
The value will deflate at some point, it's kind of a mathematical certainty. We cannot extract more money from the system than we put in, since Bitcoin is a non-productive asset. It does not generate revenue the way that a business/company does. And someone has to pay for the electricity of the miners, so money is draining out of the system, making it a negative sum game that people are playing. I would also disagree that the transactions are fast or private.
- bottlepalm 3y ago
  Private as in independent of going through any bank or institution like you would have to do to move money internationally in any other form. A store of value like gold, the dollar, or Bitcoin has nothing to do with generating revenue - only the intrinsic value of how hard is it to create more - dollars can be printed, a gold vein can be discovered tomorrow that would crash the price, Bitcoin on the other hand there's no chance of creating more than what's planned - which makes it a great store of value in turn valuable.
  - angry_albatross 3y ago
    One nice thing about gold is that we don't have to keep paying money for gold to continue to exist. It will sit on a shelf and not charge us any money for sitting there. Bitcoin exists on a network of computers that need electricity to run. If we stopped paying for this network to run, the bitcoin would stop existing, since the ownership of bitcoin really is solely determined by the ownership records on this computer network. That is my argument, that Bitcoin cannot maintain its value long term because the value is leaking out of the system in the form of electricity bills. Furthermore, we cannot let these bills become too small, or else the network becomes vulnerable to a 51% attack, so as a society we collectively must pay a large amount of "rent" on this store of value, which causes the value in this value pool to slowly deflate over time. We have been overcoming this drain so far by "investors" continuing to pour money into the system, but those investors cannot possibly get all of their money back, because it's been spent on electricity.
    - bottlepalm 3y ago
      By your own argument gold is expensive - expensive to secure, expensive to transport, slow to transport (time is money), expensive to verify. So yes it does cost money to hold and transact gold. So your same 'leaky' argument applies. Maintaining any store of value or currency is never free, but the benefits far outweigh the costs as now we have an inflation resistant medium on which to trade goods and services with.
      - angry_albatross 3y ago
        Well I'm not trying to argue for gold being a good store of value either, it might not be. But another reason why I would prefer to have gold is that people make use of gold to make jewelry and electronics, so that creates a demand that helps to prop up its value. The point I was making is that in a passive state, just sitting on a shelf, I believe that the cost of maintaining ownership of all the gold in the world is orders of magnitude lower than the cost of maintaining the entire bitcoin network, and I think that must continue to be the case due to the threat of 51% attacks. (If the cost of running the bitcoin network drops below a certain threshold, it becomes profitable for a rogue actor to rent a large amount of compute power and force in some fraudulent transactions.) So I believe that the "leaky bucket" effect is stronger with bitcoin than it is with gold, and there isn't a similar real world use case of bitcoin similar to the manufacturing of jewelry like there is for gold to counteract this leak. Therefore, the total value held by all of the holders of bitcoin must be declining due to this leak, which counteracts the idea that it is inflation resistant in the long term.
        bottlepalm 3y ago
        Your passive analogy doesn't work because Gold needs to be secured and defended, which requires energy or other people are going to steal it. Just like Bitcoin needs to be secured and defended with compute for at least new transactions. A 51% attack will allow you block new transactions or double spend coins you have, not spend or steal other people's coins on the ledger because you don't have the private keys for those.
wpietri 3y ago
Almost every word of that is untrue in practice. But let's focus on the core part, "successful". Let's compare with M-PESA, a money transfer solution that started around the same time. M-PESA has steadily grown, doing 26 billion transactions last year. [1] Bitcoin was somewhere around 100m transactions for the same period. [2] That's about 0.5% of the volume. And its worse than the raw numbers suggest, in that the M-PESA transactions were things with positive economic impact, whereas a lot of the Bitcoin activity was driven by speculation or crime. And that's wildly smaller than the number of credit and debit transactions that happened over the same time, of course. Bitcoin was hoping to be "electronic cash", but the shift away from physical cash was toward the already existing digital payment mechanisms, not Bitcoin. Merchant adoption, always small, went into decline years ago. So no, Bitcoin was not successful in the sort of terms that match its initial goals or what was hyped in the early years. [1] https://www.statista.com/statistics/1139181/m-pesa-transaction-volume/ https://www.statista.com/statistics/1139181/m-pesa-transacti... [2] based on eyeballing this: https://ycharts.com/indicators/bitcoin_transactions_per_day https://ycharts.com/indicators/bitcoin_transactions_per_day
- bottlepalm 3y ago
  Read my post, I said successful as a 'store of value' not 'electronic cash'. Bitcoin is more akin to gold - an inflation resistant store of value. There is a premium for transacting it which means you should convert it into a more inflation prone transactional currency if you want to spend it.
  - wpietri 3y ago
    Well look at those goalposts move. Bitcoin's original purpose was precisely electronic cash: https://bitcoin.org/bitcoin.pdf https://bitcoin.org/bitcoin.pdf But if you'd like to shift focus to why "store of value" is also wrong, I'm glad to. A store of value needs to be more stable than the thing you're moving out of. It also needs to be relatively liquid, and should have low transaction costs. But Bitcoin is very volatile compared with major fiat currencies, and also when compared with gold. Transaction costs are relatively high. The market is thinly traded, and is widely believed to be manipulated. Gold, in contrast, is more liquid, cheaper to trade in, and much better regulated. So no, Bitcoin doesn't make for a good store of value. People wanting to store value would generally be much better off buying index funds, which are also pretty inflation resistant, and also have a positive return. But if they want to avoid equity exposure, then the gold standard for this is, well, gold. Bitcoin is terrible by comparison.

sublinear 3y ago

It can be argued that probabilistic generative AI is about as equally worthless from first principles.

angry_albatross 3y ago
You are claiming that there is some sort of mathematical proof that this AI has no "worth"? There are a lot of people who are finding value from probabilistic generative AI, to the extent that they are willing to pay OpenAI $20/month to get access to the best model available, so I think that counteracts whatever sort of mathematical proof you might make. It's true that crypto has a lot of users too, but I think the difference is that these users are mostly being deceived into thinking that their involvement in crypto will make them rich in the future, when it will not. AI is giving people immediate value instead of making promises of future wealth.
CSMastermind 3y ago
That seems untrue to me? There are several very obvious uses for ChatGPT. Almost everyone I've seen use it can think of something they want to use it for. I've never had someone give me a plausible usecase for blockchains or crypto currencies. Once you push on the usecases I have heard it quickly devolves into "you just don't get it" or some kind of appeal to authority like "well all the VCs wouldn't be investing in it if it wasn't valuable, clearly they're smarter than you."

kayodelycaon 3y ago

The one legitimate use case of using blockchain for decentralized digital currency ends up being used by everyone who can’t transfer funds using normal routes. So most of the people using it are either bad actors or speculators Kind of like Tor. Bad actors ruin all kinds of things. People who only see technology will only go on to repeat the mistakes of the past because they lack the necessary domain knowledge to know better.

logicchains 3y ago
>So most of the people using it are either bad actors or speculators It's a matter of perspective. For many people around the world, the US government is the bad actor, the bad actor preventing them transferring funds through "normal" (US controlled) routes.
- kayodelycaon 3y ago
  Yup. The only reason blockchain has any legitimate use at all is because governments are bad actors. Doesn't change that many of the people using the system are also bad actors of one form or another.
nkuttler 3y ago
> So most of the people using it are either bad actors or speculators "bad actors" according to their government, which could mean people who simply want the right to vote, or to live free from oppression, secret police, etc.
- kayodelycaon 3y ago
  Indeed, what about all of the decent and moral people using digital currency for such reasons? Across all of them? A very low percentage. Fraud and speculation is left, right, and center. Unfortunately, only a few, like Bitcoin, can justify their existence on moral grounds. Hopefully the concept doesn’t get banned completely.
- acdha 3y ago
  Even if we take this at face value and assume these are all dissidents seeking freedom, cryptocurrencies would be a bad idea because they force you to leave a paper trail for prosecutors and deal with intermediaries who could be suborned. If Iranian dissidents buy something with cash, they might find that the person they thought was trustworthy is secretly working for the police and they’re going to have a bad time but they would at least only have proof of that single transaction. If they use cryptocurrency, the police get likely years worth of transaction history and a list of everyone you’ve worked with.

sonicshadow 3y ago

I wrong

Nextgrid 3y ago
> Plenty of great ideas and easy-to-see use cases from that first paragraph alone. Such as? Blockchains enable trust-less and decentralized ledgers at the expense of major trade-offs (user experience, lack of transaction reversibility, proof-of-work to secure the network, etc). Most useful business happens off-chain in the real-world, so you need to bridge between blockchain state and real-world state using a trusted party which throws away all the decentralization and trustlessness advantages, so you may as well just let the trusted party run a conventional database directly, and avoid the major trade-offs. I can't think of many useful & valuable use-cases that happen fully on-chain with no off-chain interactions, aside from cryptocurrencies. As soon as you have off-chain interactions (which is most of blockchain usage outside of cryptocurrency applications), the value proposition of using a blockchain goes away and a database makes more sense.

rchaud 3y ago

AI is exactly the same to marks that don't know any better and think complex new technology = $$$. AI having use cases for megacorps doesn't mean that scammers won't leverage the hype to sell get rich quick scams.

sleepybrett 3y ago

Sure but that doesn't stop the managerial and engineering tiers of those businesses to jump ship to the new hype technology. The managers didn't know shit but sales in the first place, sell coins or ai no difference. The engineers, just another new tech to learn.

tornato7 3y ago

For better or worse, Crypto is legitimately the easiest way to create a financial application, derivative, exchange etc... any programmer can create interesting, useful, and novel financial instruments like Squeeth, crvUSD, PoolTogether, etc in short order. Good luck recreating those in TradFi in under a decade, let along making them interact with each other in atomic transactions. Of course, some people see this as a bad thing because it enables scammers to create all sorts of new and improved ponzi schemes. Others think it's a good thing because it speeds up financial innovation and levels the playing field between big banks and small startups. Some might argue that crypto is only useful for building apps insofar as it avoids regulation, but you also can't convince me that Wells Fargo is the future of finance. Banks could never create a financial playground that works as well as Ethereum, even using their centralized database. IMO, so many of the projects that have come out of the crypto space are awesome and promising, but investment in the space grew faster than projects could mature. Unfortunately that leads to users losing $100M when Joe Schmo's cross-chain bridge gets hacked, when it should have never had that much TVL. Just my 2c.

arlcode 3y ago
I'd argue that the difficulty in banking is not to get the technology working. Financial software is regular software with additional audits and checks to make sure it's safe against the flood of attacks it will receive. With the kind of money we are talking about, you can hire people with the expertise to reasonably protect you against bugs and software exploits (something most traditional financial companies have a close to 100% track record in but many crypto projects failed). But after that you have a mountain of issues to consider that have little to nothing to do with software: - financial logical holes like flash loans being used to extract money "democratically" - people committing age old scams "but on the block chain" - founders not understanding problems such as that you cannot secure one unsecured coin with another one no matter the algorithm. - all sorts of unpleasant people using your "financial playground" to do things society frowns upon. All of these are solvable but not by choosing a better technology stack. Look at how much traditional institutions are spending on compliance and realise that you probably won't be able to cut that by an order of magnitude. Crypto is learning very quickly that most regulations do not exist to "keep the little man down" but because having regular people get fleeced over and over by charismatics liers/fools can have a devastating effect on any community.

RestlessMind 3y ago

> but blockchain technology was fundamentally a bad idea with no legitimate use cases. Or, you live in a bubble and haven't seen any useful examples. I have compiled some for you: https://news.ycombinator.com/item?id=32406095 https://news.ycombinator.com/item?id=32406095

angry_albatross 3y ago
These articles are mostly promoting bitcoin as a viable currency for people in desperate situations, and yes, possibly, if you are in an extreme situation like being extorted by someone with ransomware, or in a place where your government is falling apart or targeting you as a criminal, you might have some reason to use cryptocurrency in this way. But since it does such a terrible job as a currency (extremely inefficient, slow, and irreversible transactions, price volatility, no intrinsic value, etc) it should really always be a choice of last resort.