4 ms·

The author asserts that removing intermediaries (with blockchain) is a solution to a non-existent problem, as there are a lot of trusted intermediaries.

by dtech 9y ago

thieving_magpie 9y ago
I'm not knowledgeable in this area but isn't the problem "trust"? Trust is a risk. Probably a safe one but a risk nonetheless. If you have the ability to remove trust from the equation, why not pursue it?
- NathanKP 9y ago
  I think the point is that its a cost/benefit analysis. The cost of removing trust from the equation is much higher than the cost of just trusting a intermediary. In simple terms even if my bank was evil and occasionally abused my trust to steal my money it would still be cheaper to me than trying to use Bitcoin as money and paying the high transaction fees and suffering the long transaction wait times. Sure perhaps if there was no one I could trust as an intermediary and everyone was evil and would steal large amounts of money from me then a decentralized blockchain based monetary system would be worth using. But as it is the potential risk of an intermediary abusing my trust is minimal to nonexistent compared to the very real expense and wastefulness of the blockchain based systems. The same thing applies to almost all usage of blockchain. Sure it works, but trusting an intermediary just works better, and cheaper.
- SideburnsOfDoom 9y ago
  > If you have the ability to remove trust from the equation, why not pursue it? Effort vs reward. Maybe the reward is low, and the cost (e.g. in electricity to do the proof of work) is getting out of hand.
- NickM 9y ago
  We have the ability to remove it, yes, but not for free. Decentralized systems based on proof-of-work are inherently less efficient than centralized, trust-based systems. The question is, do the costs and downsides of cryptocurrencies outweigh the benefits of not having to trust a third-party intermediary? The author is arguing no, and for the overwhelming majority of real-world cases, I would tend to agree.
  - thieving_magpie 9y ago
    I understand the trade off a little better now. Thanks. It leads me to a follow up question - is this universally true of all blockchain implementations or is it due to the structure of cryptocurrency (mining, etc.)?
    - icebraining 9y ago
      Mining is part of how the Blockchain remains decentralized; you can't remove it. In the original paper, the way mining worked was by Proof-of-Work (PoW), which wastes a bunch of energy to make sure it's unprofitable to attack it. There is research into mining without wasting energy by using alternatives to PoW, like Proof-of-Stake (PoS). I don't think there's consensus about whether any of them actually work.
- hibikir 9y ago
  If one could remove the need for trust with no other costs, sure, why not? But in practice, using something shaped like bitcoin asks for very hard things: The financial system started with very little trust, but it expanded because trust solved problems, like money being stolen on route, an associate running away and leaving and bank runs. Trust makes those problems mostly go away. Bitcoin has digital equivalents: 51% attacks, someone duplicating a private key, someone losing their private key and the money disappearing, exchanges becoming insolvent, getting hacked, or downright stealing, low transaction throughput, high costs per transaction, and high volatility. If you study the history of banking and finance, the blockchain is in 1700-1800 or so, and we solved those problems through more trust and faster communication. Our best solutions to both the blockchain and old finance problems? More trust, as it lowered overall risk. We can't just improve the fact that the current system requires a lot of trust with a new system that just happens to still need trust to work efficiently. Now, maybe we'll solve those problems in ways that don't requite more trust, but until then, cryptocurrencies will we far riskier than the existing financial system.
- conanbatt 9y ago
  Yes but blockchain also introduced a new risk, or magnified it. Your funds can get stolen without you knowing about it and there's nothing you can do. That cannot happen with a bank account.
- JustAnotherPat 9y ago
  Trust acts as capital in today's market. You can often get some sort of payback if you trust somebody else with your money.
- dithering 9y ago
  Suggestion: (needs trust + fallback to a court of law) is less risky than (trust free + no fallback). It's a moot point anyway - as has been said, bitcoin's a commodity not a currency.
- bunderbunder 9y ago
  If you do, sure. But nobody has that ability. I just mentioned, elsewhere in this thread, a couple major players in the cryptocurrency space who have proved to be untrustworthy. And a lot of people got burned in part because they believed the hype that removing the need for trust from the management of the "physical" currency is equivalent to removing the need for trust from the financial system itself. That might have been true thousands of years ago, before the birth of finance, when everything really did run on just cash. Things are more complicated than that nowadays, though. Cash is such a small corner of the money supply in a modern economy. A few percent of the bits that people bother to try and count, and there's a lot more that goes uncounted. A distributed trustless public ledger for the cash doesn't really help with the trust issues for that stuff. Frankly, it only helps with the small subset of money for which I'm least worried about trust.
viggity 9y ago
Yes, there are a lot of trusted intermediaries if you live in the US, Europe, or otherwise sufficiently developed countries. However, there are few if any truly trusted intermediaries in the developing world (and countries that have regressed significantly), including India, Venezuela, Zimbabwe. Perhaps blockchain is superior to a trusted intermediary, perhaps not. But it isn't gonna take off in the developed world until it proves its utility in the undeveloped world.
xwvvvvwx 9y ago
We've been automating middlemen away for years, don't see whats different here? Blockchain lets us do some stuff now that was impossible before (single ownership of digital assets jumps out), but a big chunk of the economic benefit is going to come from reduced transaction costs.
- conanbatt 9y ago
  The author makes the point that the proof-of-work is expensive enough to have the transaction costs still too high to compete with the current financial environment. I do not send money to my wife on bitcoin, I use venmo or bank accounts and it is free and immediate.
- dtech 9y ago
  Bitcoin, the only "proven" crypto coin thus far, totally failed as a usable currency because of the high transaction costs. Other cryptocoins including BTC, Ethereum, Litecoin or whatever do not have lower transaction cost than traditional currency. In fact I would argue that since public blockchain require mining transaction costs can never be that much lower than traditional currency.
- Asdfbla 9y ago
  And blockchains achieve that at an immense cost, which is the point of the discussion. Decentralised systems are less efficient (orders of magnitude less for proof of work) and it's debatable whether that is necessary in the real world, where trust exists for the majority of applications.
- IkmoIkmo 9y ago
  We've been reducing middlemen to few, not to zero. e.g. Amazon, Whatsapp, Facebook, Airbnb etc. Hell, even peer-to-peer systems have some kind of middlemen. If that middleman is in a big enough market with low barriers to entry and lots of competition, and is able to drive cost down with smart tech, it doesn't really matter to most users whether he exists or not. Decentralised systems never promised low transaction costs. A private party running a database can do it cheaper and faster than a set of decentralised actors working for some incentive to establish the same database. That seems pretty much inevitable. They did promise something else. Trustless-systems. But it hasn't been shown yet that consumers value this. (e.g. see our most valued companies like Facebook to which we routinely trust data (from secrets to nudes) we'd never, ever share with anyone else) Blockchain's value proposition, as I've always interpreted it, isn't 'material' in the sense that transaction costs or speed are, but more based on the culture and the relationships of the 'value network'. i.e., the relationships aren't based on trust+laws, but on code+math. And it doesn't appear to me that the market values these relationships/culture, but it seems that the market does value the material aspects. hence there being no real blockchain users of blockchain applications. (except for the whole store of value thing, which drives basically all the price changes, but nobody envisioned the blockchain to solely be digital gold and nothing else).
xutopia 9y ago
If I told you that Equifax is one of those "trusted intermediaries"...

The author asserts that removing intermediaries (with blockchain) is a solution to a non-existent problem, as there are a lot of trusted intermediaries.

by dtech 9y ago

The author asserts that removing intermediaries (with blockchain) is a solution to a non-existent problem, as there are a lot of trusted intermediaries.

thieving_magpie 9y ago

I'm not knowledgeable in this area but isn't the problem "trust"? Trust is a risk. Probably a safe one but a risk nonetheless. If you have the ability to remove trust from the equation, why not pursue it?

NathanKP 9y ago
I think the point is that its a cost/benefit analysis. The cost of removing trust from the equation is much higher than the cost of just trusting a intermediary. In simple terms even if my bank was evil and occasionally abused my trust to steal my money it would still be cheaper to me than trying to use Bitcoin as money and paying the high transaction fees and suffering the long transaction wait times. Sure perhaps if there was no one I could trust as an intermediary and everyone was evil and would steal large amounts of money from me then a decentralized blockchain based monetary system would be worth using. But as it is the potential risk of an intermediary abusing my trust is minimal to nonexistent compared to the very real expense and wastefulness of the blockchain based systems. The same thing applies to almost all usage of blockchain. Sure it works, but trusting an intermediary just works better, and cheaper.
SideburnsOfDoom 9y ago
> If you have the ability to remove trust from the equation, why not pursue it? Effort vs reward. Maybe the reward is low, and the cost (e.g. in electricity to do the proof of work) is getting out of hand.
NickM 9y ago
We have the ability to remove it, yes, but not for free. Decentralized systems based on proof-of-work are inherently less efficient than centralized, trust-based systems. The question is, do the costs and downsides of cryptocurrencies outweigh the benefits of not having to trust a third-party intermediary? The author is arguing no, and for the overwhelming majority of real-world cases, I would tend to agree.
- thieving_magpie 9y ago
  I understand the trade off a little better now. Thanks. It leads me to a follow up question - is this universally true of all blockchain implementations or is it due to the structure of cryptocurrency (mining, etc.)?
  - icebraining 9y ago
    Mining is part of how the Blockchain remains decentralized; you can't remove it. In the original paper, the way mining worked was by Proof-of-Work (PoW), which wastes a bunch of energy to make sure it's unprofitable to attack it. There is research into mining without wasting energy by using alternatives to PoW, like Proof-of-Stake (PoS). I don't think there's consensus about whether any of them actually work.
hibikir 9y ago
If one could remove the need for trust with no other costs, sure, why not? But in practice, using something shaped like bitcoin asks for very hard things: The financial system started with very little trust, but it expanded because trust solved problems, like money being stolen on route, an associate running away and leaving and bank runs. Trust makes those problems mostly go away. Bitcoin has digital equivalents: 51% attacks, someone duplicating a private key, someone losing their private key and the money disappearing, exchanges becoming insolvent, getting hacked, or downright stealing, low transaction throughput, high costs per transaction, and high volatility. If you study the history of banking and finance, the blockchain is in 1700-1800 or so, and we solved those problems through more trust and faster communication. Our best solutions to both the blockchain and old finance problems? More trust, as it lowered overall risk. We can't just improve the fact that the current system requires a lot of trust with a new system that just happens to still need trust to work efficiently. Now, maybe we'll solve those problems in ways that don't requite more trust, but until then, cryptocurrencies will we far riskier than the existing financial system.
conanbatt 9y ago
Yes but blockchain also introduced a new risk, or magnified it. Your funds can get stolen without you knowing about it and there's nothing you can do. That cannot happen with a bank account.
JustAnotherPat 9y ago
Trust acts as capital in today's market. You can often get some sort of payback if you trust somebody else with your money.
dithering 9y ago
Suggestion: (needs trust + fallback to a court of law) is less risky than (trust free + no fallback). It's a moot point anyway - as has been said, bitcoin's a commodity not a currency.
bunderbunder 9y ago
If you do, sure. But nobody has that ability. I just mentioned, elsewhere in this thread, a couple major players in the cryptocurrency space who have proved to be untrustworthy. And a lot of people got burned in part because they believed the hype that removing the need for trust from the management of the "physical" currency is equivalent to removing the need for trust from the financial system itself. That might have been true thousands of years ago, before the birth of finance, when everything really did run on just cash. Things are more complicated than that nowadays, though. Cash is such a small corner of the money supply in a modern economy. A few percent of the bits that people bother to try and count, and there's a lot more that goes uncounted. A distributed trustless public ledger for the cash doesn't really help with the trust issues for that stuff. Frankly, it only helps with the small subset of money for which I'm least worried about trust.

viggity 9y ago

Yes, there are a lot of trusted intermediaries if you live in the US, Europe, or otherwise sufficiently developed countries. However, there are few if any truly trusted intermediaries in the developing world (and countries that have regressed significantly), including India, Venezuela, Zimbabwe. Perhaps blockchain is superior to a trusted intermediary, perhaps not. But it isn't gonna take off in the developed world until it proves its utility in the undeveloped world.

xwvvvvwx 9y ago

We've been automating middlemen away for years, don't see whats different here? Blockchain lets us do some stuff now that was impossible before (single ownership of digital assets jumps out), but a big chunk of the economic benefit is going to come from reduced transaction costs.

conanbatt 9y ago
The author makes the point that the proof-of-work is expensive enough to have the transaction costs still too high to compete with the current financial environment. I do not send money to my wife on bitcoin, I use venmo or bank accounts and it is free and immediate.
dtech 9y ago
Bitcoin, the only "proven" crypto coin thus far, totally failed as a usable currency because of the high transaction costs. Other cryptocoins including BTC, Ethereum, Litecoin or whatever do not have lower transaction cost than traditional currency. In fact I would argue that since public blockchain require mining transaction costs can never be that much lower than traditional currency.
Asdfbla 9y ago
And blockchains achieve that at an immense cost, which is the point of the discussion. Decentralised systems are less efficient (orders of magnitude less for proof of work) and it's debatable whether that is necessary in the real world, where trust exists for the majority of applications.
IkmoIkmo 9y ago
We've been reducing middlemen to few, not to zero. e.g. Amazon, Whatsapp, Facebook, Airbnb etc. Hell, even peer-to-peer systems have some kind of middlemen. If that middleman is in a big enough market with low barriers to entry and lots of competition, and is able to drive cost down with smart tech, it doesn't really matter to most users whether he exists or not. Decentralised systems never promised low transaction costs. A private party running a database can do it cheaper and faster than a set of decentralised actors working for some incentive to establish the same database. That seems pretty much inevitable. They did promise something else. Trustless-systems. But it hasn't been shown yet that consumers value this. (e.g. see our most valued companies like Facebook to which we routinely trust data (from secrets to nudes) we'd never, ever share with anyone else) Blockchain's value proposition, as I've always interpreted it, isn't 'material' in the sense that transaction costs or speed are, but more based on the culture and the relationships of the 'value network'. i.e., the relationships aren't based on trust+laws, but on code+math. And it doesn't appear to me that the market values these relationships/culture, but it seems that the market does value the material aspects. hence there being no real blockchain users of blockchain applications. (except for the whole store of value thing, which drives basically all the price changes, but nobody envisioned the blockchain to solely be digital gold and nothing else).

xutopia 9y ago

If I told you that Equifax is one of those "trusted intermediaries"...