5 ms·

I think it's fair to say that most tokens are effectively trash, and there are many scams in crypto. Relative to other speculative manias in history, crypto is

by spir 4y ago

I think it's fair to say that most tokens are effectively trash, and there are many scams in crypto. Relative to other speculative manias in history, crypto is a mechanism for mania and speculation. You can't trade tulips on tulips. But you can trade crypto on crypto. For example, in the British railway mania of the 1840s, many patterns occurred similar to today's crypto frenzy, such as stock exchanges created expressly for the purpose of trading railway stocks and publications to talk about and advertise railway stocks. But railways lack crypto's reflexivity because you can't trade railway stocks on railways. Crypto mechanizes its own speculative mania. But what's the other side here? Does it even exist? I think it does exist. In my view, a balanced discussion of the pros and cons of crypto often seems to elude the HN community. HN is a community of technologists, right? So what's the actual technology here? At the root of crypto are two technologies, 1) programmatic public chains that are inexpensive to run and 2) zero-knowledge proofs. By combining public chains and zero-knowledge proofs, we get an inherently global market, with p2p transactions that scale to all of humanity, where you can send money digitally, similar to handing a $20 bill to a friend. And the transfers can include rigid, sophisticated logic that offers the potential to reduce transaction costs for many kinds of routine economic activity. So, while it's fair to say that most crypto tokens are effectively trash and crypto mechanizes scams and speculation, it's also unfair to omit discussion of the fundamental innovation, and those who do so will end up on the wrong side of history.

leonidasv 4y ago
The problem with most approaches is that they throw out the baby with the bathwater. Sure, you trusting third-parties is shady. Sure, having governments handling over your money is bad. However, having a fully-immutable chain is not the solution. Let's say, you tokenized the ownership of your home but a bad actor from the other side of the globe stole the token. A standard certificate of ownership can still be brought to court in case of fraud, but a token cannot. I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. So you'll end-up with a token that does not represent the reality - and that's why standard, traditional certificates exist and are trusted: they represent the rule of the law, something an immutable decentralized chain can't do. Another example is just what is happening to Russia and its oligarchs. Traditional banks can be forced by the rule of the law to impose sanctions; crypto cannot. So they're just evading those sanctions in plain sight by using crypto. You may side with Russia on the current affairs, but this also applies to terrorist groups, gangs, you name it. And it goes on and on... if you're a victim of a phishing attack like the one happened recently at OpenSea[1], but instead of signing with your wallet you filled your credit card data into a shady field, you can still contest the purchase with your bank and have your money back. Even if you authorized a one-way money transfer from your bank account, in some countries you can still contest it and eventually have your money back.[2] None of this is available for the victims of the NFT phishing scam. So, while the technology sure has interesting points, the current use-cases are far from optimal. Society developed trust and trust-chains for a reason. If you're a going to implement this in blockchain, then you'll have to build a non-decentralized* approach from the ground up, that most of the crypto community will bash you for. * To be fair, many consider the proof-of-stake to be a non-decentralized approach departing from the vision of S. Nakamoto. And some will even argue that the current proof-of-work mining concentration is already centralized.[3] [1] https://cointelegraph.com/news/opensea-phishing-scandal-reveals-a-security-need-across-the-nft-landscape https://cointelegraph.com/news/opensea-phishing-scandal-reve... [2] https://www-poder360-com-br.translate.goog/justica/globo-ganha-na-justica-casa-paga-com-pix-errado-de-r-318-mil/?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp https://www-poder360-com-br.translate.goog/justica/globo-gan... [3] https://www.technologyreview.com/2022/03/04/1046636/ethereum-blockchain-proof-of-stake/ https://www.technologyreview.com/2022/03/04/1046636/ethereum...
- logifail 4y ago
  > I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. That's within a whisker of saying that the entire legal system is subservient to blockchain, and not the other way around. If so it would appear to be the death-knell for all those use-cases for any official/governmental process to use blockchain... ?
  - throwaway82652 4y ago
    There never was a chance for them to be used in an official/governmental process because they don't offer anything there. Those processes require a trusted authority to work correctly. Why would they bother "decentralizing" something like this to put it on a blockchain, entrusting it to some unknown third parties to run the network? There's no point to that at all.
- saurik 4y ago
  > I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. So you'll end-up with a token that does not represent the reality - and that's why standard, traditional certificates exist and are trusted: they represent the rule of the law, something an immutable decentralized chain can't do. Unless the token is being used to control some kind of elaborate--and likely magical--boobytrap (to enforce that the owner of the token is the only person who can even step foot on the property), the government would just say that old token was no longer authoritative and that some new token is... which is no different from how a standard title or deed also can't be blindly trusted (as it may have been revoked by the government). And yet, forging your sheets of paper or invalid transfer signatures on the real sheet of paper (how classic titles and deeds work, as in when I signed over ownership of a car once) is way easier than forging this blockchain record, and so this actually still provides a ton of benefit by avoiding cases that might have to go to court in the first place.
  - throwaway82652 4y ago
gigaflop 4y ago
I agree that most tokens are scammy trash, especially when advertised, but acknowledge that some will stand the test of time. I don't think I'll ever hook up a wallet to my smartphone in order to use tap-to-pay in some hypothetical future, but I really do believe that crypto can be great at hopping traditional regulatory fences. If I come across some desirable service offered on the web by a guy in Namibia, they're much more likely to get money from me if they accept something that I can easily acquire and send. I don't even know what Namibia's currency is, but they can list a price in BTC/ETH/DOGE and I can easily make my own judgements on whether it's worth it for me.
przeor 4y ago
I agree. In dot com era most of startups are not here? Only circa1%~3% of founded between 1995 to 2001 created real economic value. Does it mean rest 97% were scams because they didnt create meaningful value etc.?
- AlexandrB 4y ago
  I think many of them were, yes. Not in the sense that the company founders we're trying to scam investors but in the sense that VCs and the banks involved in their IPOs hyped them up and dumped them on the public market despite knowing how little of substance was going on. Basically the people that should have known better were happy to promote garbage to those who don't know better. The same thing happened again in 2008 with mortgage derivatives and now I think we're headed for round 3 with crypto.
  - przeor 4y ago
    agree
cslarson 4y ago
Ha this is so well put, thank you. But it does surprise that so many technologists are falling into this trap (disregarding the innovation because of the plethora of junk it enables). The value of permission-less platforms and protocols should no be assessed absent of curation. Google, email spam filtering, wikipedia, the web abounds with examples and without curation we would be in a sea of shit.
- AlexandrB 4y ago
  Aren't these examples of centralization of a decentralized technology? If so, what's the point of starting an expensive-to-run distributed platform only to end up at the same point we're at now where in practice most activity goes through a few centralized gatekeepers? After all, you can still run your own web or SMTP server. The web is still permission-less in the trivial sense - just like crypto. But Google can blacklist your site or blackhole your emails making your permission-less services useless in practice. What stops a curated web3 from ending up in the same place we're at now?
  - throwaway82652 4y ago
    >What stops a curated web3 from ending up in the same place we're at now? The answer is either nothing at all, or paradoxically it's some vague sense of optimism involving "faith" "belief" and "trust". It should surprise no one that web3 fundamentally can't deliver on any of its promises.
throwaway82652 4y ago
>it's also unfair to omit discussion of the fundamental innovation, and those who do so will end up on the wrong side of history. No, please stop with this fallacy. I don't know where people get this idea besides from marketing hype. There is no innovation in cryptocurrency. It's perfectly fair to also disregard any claims of innovation, because they're not true. I'll go into the specifics here. >1) programmatic public chains that are inexpensive to run The innovation you're taking about here is a database server, blockchains not necessary. AWS and Azure are also an "inexpensive programmable and public" server except you can put anything you want on it, it doesn't have to take the form of an immutable ledger. Blockchains are just strictly worse than everything else. >2) zero-knowledge proofs. Also has nothing to do with blockchains, zero knowledge proofs are useful for other purposes besides cryptocurrency, and were invented long before it. >By combining public chains and zero-knowledge proofs, we get an inherently global market, with p2p transactions that scale to all of humanity, where you can send money digitally, similar to handing a $20 bill to a friend. No, you don't. By combining those things you might get some structure that would theoretically allow you to do that, but that's only a small portion of building a scalable financial platform. There are still a ton of other services you have to build, which have to exist anyway for any form of online finance regardless of whether you use a blockchain or not. Blockchains are also not "p2p" in any way, by definition they require middlemen to run the chain. I'm growing tired of reading these comments about cryptocurrency where basically everything is wrong, it's exhausting to deal with the torrent of misleading statements. That should tip any technologist off that something is seriously wrong there. I can't even blame you for making them, the marketing hype is just out of control and is confusing everybody.
- robot9000 4y ago
  >There is no innovation in cryptocurrency What other technology makes use of an immutable ledger, like you said?
  - davidgerard 4y ago
    Merkle trees were invented in 1979.
- 4y ago
dralley 4y ago
>programmatic public chains that are inexpensive to run If you consider gas fees as a cost for running the chain, they absolutely are not inexpensive to run. They may well be the most expensive ledger on earth. Writing the numbers down on-per-page in a book made of gold leaf would probably be cheaper.
- spir 4y ago
  You're right that ethereum transaction fees are very high (https://cryptofees.info/ https://cryptofees.info/). imo, total gas fees seem likely to continue to hyper-grow. Yet, per-transaction or per-user gas fees may be expected to plummet due to scaling technologies, especially https://arbitrum.io https://arbitrum.io, https://optimism.io https://optimism.io, and https://starknet.io https://starknet.io (they do like those io's)
  - rmbyrro 4y ago
    Maybe because everybody else like those .com's and none of them are left anymore...
davidgerard 4y ago
> So what's the actual technology here? Shared Merkle tree ledgers are the technology that actually exists, and has done since the 1990s. It turns out to have very few applications. https://blog.dshr.org/2022/02/ee380-talk.html https://blog.dshr.org/2022/02/ee380-talk.html - a talk by David Rosenthal, who made a few things using shared Merkle tree ledgers in the 1990s and early 2000s. The rest of the "technology" is vague promises and get-rich-quick schemes, using technology as the excuse. e.g., you're talking about a "global market" using ZKPs, but no such thing exists - you're stating a "blockchain could" as a "blockchain does". When "could" is a word that means "doesn't."
dmitriid 4y ago
> In my view, a balanced discussion of the pros and cons of crypto often seems to elude the HN community. No. It eludes the crypto community. For every well-researched article pointing out flaws, there are exactly zero articles addressing any points in those articles. There's no end to "just join the discords and read up" though. > At the root of crypto are two technologies, 1) programmatic public chains that are inexpensive to run Those chains are extremely expensive to run as evidenced by the absolutely dominant one. > 2) zero-knowledge proofs. These are not at the "root fo crypto". This is something tacked on to try and solve the 0.00001% of crypto's problems. > By combining public chains and zero-knowledge proofs, we get an inherently global market, with p2p transactions that scale to all of humanity No, no we don't. > it's also unfair to omit discussion of the fundamental innovation, and those who do so will end up on the wrong side of history. The proof is in the pudding as it were. There's literally nothing in this "fundamental innovation" that can't be done better, faster, cheaper, and on a bigger scale with literally anything else. - Blockchains are distributed immutable ledgers, and there are very few (if any) useful applications for them - "programmatic chains" are the worlds slowest, most ineffecient and expensive VM that gets exploited every day because even people programming it don't understand the rules they are writing
JoachimS 4y ago
"programmatic public chains that are inexpensive to run" Even if we ignore the equipment and operational cost, for example the energy cost of mining (dump it on a third party for example), you can't still claim that it is inexpensive to run. At least not without some serious numbers backing up the claim.

by spir 4y ago

leonidasv 4y ago

The problem with most approaches is that they throw out the baby with the bathwater. Sure, you trusting third-parties is shady. Sure, having governments handling over your money is bad. However, having a fully-immutable chain is not the solution. Let's say, you tokenized the ownership of your home but a bad actor from the other side of the globe stole the token. A standard certificate of ownership can still be brought to court in case of fraud, but a token cannot. I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. So you'll end-up with a token that does not represent the reality - and that's why standard, traditional certificates exist and are trusted: they represent the rule of the law, something an immutable decentralized chain can't do. Another example is just what is happening to Russia and its oligarchs. Traditional banks can be forced by the rule of the law to impose sanctions; crypto cannot. So they're just evading those sanctions in plain sight by using crypto. You may side with Russia on the current affairs, but this also applies to terrorist groups, gangs, you name it. And it goes on and on... if you're a victim of a phishing attack like the one happened recently at OpenSea[1], but instead of signing with your wallet you filled your credit card data into a shady field, you can still contest the purchase with your bank and have your money back. Even if you authorized a one-way money transfer from your bank account, in some countries you can still contest it and eventually have your money back.[2] None of this is available for the victims of the NFT phishing scam. So, while the technology sure has interesting points, the current use-cases are far from optimal. Society developed trust and trust-chains for a reason. If you're a going to implement this in blockchain, then you'll have to build a non-decentralized* approach from the ground up, that most of the crypto community will bash you for. * To be fair, many consider the proof-of-stake to be a non-decentralized approach departing from the vision of S. Nakamoto. And some will even argue that the current proof-of-work mining concentration is already centralized.[3] [1] https://cointelegraph.com/news/opensea-phishing-scandal-reveals-a-security-need-across-the-nft-landscape https://cointelegraph.com/news/opensea-phishing-scandal-reve... [2] https://www-poder360-com-br.translate.goog/justica/globo-ganha-na-justica-casa-paga-com-pix-errado-de-r-318-mil/?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp https://www-poder360-com-br.translate.goog/justica/globo-gan... [3] https://www.technologyreview.com/2022/03/04/1046636/ethereum-blockchain-proof-of-stake/ https://www.technologyreview.com/2022/03/04/1046636/ethereum...

logifail 4y ago
> I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. That's within a whisker of saying that the entire legal system is subservient to blockchain, and not the other way around. If so it would appear to be the death-knell for all those use-cases for any official/governmental process to use blockchain... ?
- throwaway82652 4y ago
  There never was a chance for them to be used in an official/governmental process because they don't offer anything there. Those processes require a trusted authority to work correctly. Why would they bother "decentralizing" something like this to put it on a blockchain, entrusting it to some unknown third parties to run the network? There's no point to that at all.
saurik 4y ago
> I mean, you still can bring it to court, have the judge decide in your favor, but the chain is immutable and neither you or the court officials can do anything to revert that. So you'll end-up with a token that does not represent the reality - and that's why standard, traditional certificates exist and are trusted: they represent the rule of the law, something an immutable decentralized chain can't do. Unless the token is being used to control some kind of elaborate--and likely magical--boobytrap (to enforce that the owner of the token is the only person who can even step foot on the property), the government would just say that old token was no longer authoritative and that some new token is... which is no different from how a standard title or deed also can't be blindly trusted (as it may have been revoked by the government). And yet, forging your sheets of paper or invalid transfer signatures on the real sheet of paper (how classic titles and deeds work, as in when I signed over ownership of a car once) is way easier than forging this blockchain record, and so this actually still provides a ton of benefit by avoiding cases that might have to go to court in the first place.
- throwaway82652 4y ago

gigaflop 4y ago

I agree that most tokens are scammy trash, especially when advertised, but acknowledge that some will stand the test of time. I don't think I'll ever hook up a wallet to my smartphone in order to use tap-to-pay in some hypothetical future, but I really do believe that crypto can be great at hopping traditional regulatory fences. If I come across some desirable service offered on the web by a guy in Namibia, they're much more likely to get money from me if they accept something that I can easily acquire and send. I don't even know what Namibia's currency is, but they can list a price in BTC/ETH/DOGE and I can easily make my own judgements on whether it's worth it for me.

przeor 4y ago

I agree. In dot com era most of startups are not here? Only circa1%~3% of founded between 1995 to 2001 created real economic value. Does it mean rest 97% were scams because they didnt create meaningful value etc.?

AlexandrB 4y ago
I think many of them were, yes. Not in the sense that the company founders we're trying to scam investors but in the sense that VCs and the banks involved in their IPOs hyped them up and dumped them on the public market despite knowing how little of substance was going on. Basically the people that should have known better were happy to promote garbage to those who don't know better. The same thing happened again in 2008 with mortgage derivatives and now I think we're headed for round 3 with crypto.
- przeor 4y ago
  agree

cslarson 4y ago

Ha this is so well put, thank you. But it does surprise that so many technologists are falling into this trap (disregarding the innovation because of the plethora of junk it enables). The value of permission-less platforms and protocols should no be assessed absent of curation. Google, email spam filtering, wikipedia, the web abounds with examples and without curation we would be in a sea of shit.

AlexandrB 4y ago
Aren't these examples of centralization of a decentralized technology? If so, what's the point of starting an expensive-to-run distributed platform only to end up at the same point we're at now where in practice most activity goes through a few centralized gatekeepers? After all, you can still run your own web or SMTP server. The web is still permission-less in the trivial sense - just like crypto. But Google can blacklist your site or blackhole your emails making your permission-less services useless in practice. What stops a curated web3 from ending up in the same place we're at now?
- throwaway82652 4y ago
  >What stops a curated web3 from ending up in the same place we're at now? The answer is either nothing at all, or paradoxically it's some vague sense of optimism involving "faith" "belief" and "trust". It should surprise no one that web3 fundamentally can't deliver on any of its promises.

throwaway82652 4y ago

>it's also unfair to omit discussion of the fundamental innovation, and those who do so will end up on the wrong side of history. No, please stop with this fallacy. I don't know where people get this idea besides from marketing hype. There is no innovation in cryptocurrency. It's perfectly fair to also disregard any claims of innovation, because they're not true. I'll go into the specifics here. >1) programmatic public chains that are inexpensive to run The innovation you're taking about here is a database server, blockchains not necessary. AWS and Azure are also an "inexpensive programmable and public" server except you can put anything you want on it, it doesn't have to take the form of an immutable ledger. Blockchains are just strictly worse than everything else. >2) zero-knowledge proofs. Also has nothing to do with blockchains, zero knowledge proofs are useful for other purposes besides cryptocurrency, and were invented long before it. >By combining public chains and zero-knowledge proofs, we get an inherently global market, with p2p transactions that scale to all of humanity, where you can send money digitally, similar to handing a $20 bill to a friend. No, you don't. By combining those things you might get some structure that would theoretically allow you to do that, but that's only a small portion of building a scalable financial platform. There are still a ton of other services you have to build, which have to exist anyway for any form of online finance regardless of whether you use a blockchain or not. Blockchains are also not "p2p" in any way, by definition they require middlemen to run the chain. I'm growing tired of reading these comments about cryptocurrency where basically everything is wrong, it's exhausting to deal with the torrent of misleading statements. That should tip any technologist off that something is seriously wrong there. I can't even blame you for making them, the marketing hype is just out of control and is confusing everybody.

robot9000 4y ago
>There is no innovation in cryptocurrency What other technology makes use of an immutable ledger, like you said?
- davidgerard 4y ago
  Merkle trees were invented in 1979.
4y ago

dralley 4y ago

>programmatic public chains that are inexpensive to run If you consider gas fees as a cost for running the chain, they absolutely are not inexpensive to run. They may well be the most expensive ledger on earth. Writing the numbers down on-per-page in a book made of gold leaf would probably be cheaper.

spir 4y ago
You're right that ethereum transaction fees are very high (https://cryptofees.info/ https://cryptofees.info/). imo, total gas fees seem likely to continue to hyper-grow. Yet, per-transaction or per-user gas fees may be expected to plummet due to scaling technologies, especially https://arbitrum.io https://arbitrum.io, https://optimism.io https://optimism.io, and https://starknet.io https://starknet.io (they do like those io's)
- rmbyrro 4y ago
  Maybe because everybody else like those .com's and none of them are left anymore...

davidgerard 4y ago

> So what's the actual technology here? Shared Merkle tree ledgers are the technology that actually exists, and has done since the 1990s. It turns out to have very few applications. https://blog.dshr.org/2022/02/ee380-talk.html https://blog.dshr.org/2022/02/ee380-talk.html - a talk by David Rosenthal, who made a few things using shared Merkle tree ledgers in the 1990s and early 2000s. The rest of the "technology" is vague promises and get-rich-quick schemes, using technology as the excuse. e.g., you're talking about a "global market" using ZKPs, but no such thing exists - you're stating a "blockchain could" as a "blockchain does". When "could" is a word that means "doesn't."

dmitriid 4y ago

> In my view, a balanced discussion of the pros and cons of crypto often seems to elude the HN community. No. It eludes the crypto community. For every well-researched article pointing out flaws, there are exactly zero articles addressing any points in those articles. There's no end to "just join the discords and read up" though. > At the root of crypto are two technologies, 1) programmatic public chains that are inexpensive to run Those chains are extremely expensive to run as evidenced by the absolutely dominant one. > 2) zero-knowledge proofs. These are not at the "root fo crypto". This is something tacked on to try and solve the 0.00001% of crypto's problems. > By combining public chains and zero-knowledge proofs, we get an inherently global market, with p2p transactions that scale to all of humanity No, no we don't. > it's also unfair to omit discussion of the fundamental innovation, and those who do so will end up on the wrong side of history. The proof is in the pudding as it were. There's literally nothing in this "fundamental innovation" that can't be done better, faster, cheaper, and on a bigger scale with literally anything else. - Blockchains are distributed immutable ledgers, and there are very few (if any) useful applications for them - "programmatic chains" are the worlds slowest, most ineffecient and expensive VM that gets exploited every day because even people programming it don't understand the rules they are writing

JoachimS 4y ago

"programmatic public chains that are inexpensive to run" Even if we ignore the equipment and operational cost, for example the energy cost of mining (dump it on a third party for example), you can't still claim that it is inexpensive to run. At least not without some serious numbers backing up the claim.