The crypto thread

[MrCynical]

I guess there are questions about the magnitude of risk of losing your key/seed to a fire or rubber hose decryption or whatever compared to bank scams, but there is a lot of room for improvement.

Crypto seems to require choosing between keeping your key in physical form, thereby getting you all the vulnerabilities of the "cash in the mattress" approach, and keeping it purely in your head. Which is where the perennial news stories of "I have a fortune in Bitcoin but since I've forgotten the password it's worthless because there are literally no recovery options" come from. And that's before we get onto the technical side of what software you're actually putting that key into, and the blockchain behind it. I don't pretend to be an expert on any of that, which is why I recognize it would be stupid for me to take sole responsibility for all security issues there.

Of course if you want to play with smart contracts and trying to make money with speculation that is going to be a lot harder. I am not at all sure I have seen a good use case for that frenzy of speculation.

Ah yes, smart contracts. Another example that bad ideas are disproportionately likely to have the word "smart" in the name. I have just enough experience coding, and having code behave in ways other than I intended, to find it horrifyingly hilarious when people say "the code is the law" like that's a good thing...

 

[Samson]

Crypto seems to require choosing between keeping your key in physical form, thereby getting you all the vulnerabilities of the "cash in the mattress" approach, and keeping it purely in your head. Which is where the perennial news stories of "I have a fortune in Bitcoin but since I've forgotten the password it's worthless because there are literally no recovery options" come from.

It is certainly a problem. There are halfway houses between the two extremes, eg. write down the seed but remember that you have to add your birth date or something.

Ah yes, smart contracts. Another example that bad ideas are disproportionately likely to have the word "smart" in the name. I have just enough experience coding, and having code behave in ways other than I intended, to find it horrifyingly hilarious when people say "the code is the law" like that's a good thing...

I know little about them, but it seems to me they are missing a test platform to run a smart contract on to see what it does. As it is people use wallets (metamask) that will run arbitrary code using your keys and people do not thoroughly test the code first

 

[warpus]

How does blockchain tech help, well, anything, in centralised finance? It's little more than a really inefficient database.

When I make a credit card payment it takes like 3-5 business days to register. When my parents send money overseas, it's crazy expensive and takes a while to arrive too. You gotta jump through a lot of hoops.

These are just two things the blockchain is better at than a centralized banking system. Until the banks got off their butts and improve things a bit that is. But I suppose they don't really have any motivation to, for now

The blockchain is also 100% transparent. It's a lot harder the perpetrate white collar crime on it since every single transaction is public. No, it doesn't remove bad actors from the system, but it opens everything up for everyone to see, which helps.

Other than that I'm not really sure, tbh. Our centralized banking system seems way overdue for an upgrade though, don't you think? Payments and transactions shouldn't take days to clear (or longer). Why does nothing ever happen on weekends? Or at night? It should all be pretty much instantaneous and available to us 24/7. Isn't this supposed to be the information age? The banks live in the past.

Let me also note that if banks and other financial institutions didn't think they could use blockchain tech to improve their services, they wouldn't be buying in and incorporating the tech the way they are right now. As with any technology, there are likely improvements there that aren't obvious yet.

You could say that the blockchain is nothing more than a decentralized ledger. You could also say that electricity is nothing more than moving electrons though. The implications of these technologies require a lot more digging and research than just a cursory glance at the surface.

Sounds like crypto is only really useful for criminals or the paranoid.

I mean, it is being used for all sorts of things right now, including, for example, helping companies clean up our oceans. Here is one example. There are more.

A programmable decentralized ledger can be used for a lot more than just finance related projects. And it is.

 

[TheCleanerDragon]

I mean, it is being used for all sorts of things right now, including, for example, helping companies clean up our oceans. Here is one example. There are more.

A programmable decentralized ledger can be used for a lot more than just finance related projects. And it is.

That example doesn't seem to do anything that couldn't be done before crypto was invented. Their main gimmick, Sea Chain Token, is functionally identical to stock.

 

[MrCynical]

I mean, it is being used for all sorts of things right now, including, for example, helping companies clean up our oceans. Here is one example. There are more.

I love the sheer amount of wordy fluff they've deployed to hide the gaping chasm in the plan between "buy into our particular crypto-token" and "sea pollution is reduced - somehow..."

 

[PhroX]

When I make a credit card payment it takes like 3-5 business days to register. When my parents send money overseas, it's crazy expensive and takes a while to arrive too. You gotta jump through a lot of hoops.

These are just two things the blockchain is better at than a centralized banking system. Until the banks got off their butts and improve things a bit that is. But I suppose they don't really have any motivation to, for now

The blockchain is also 100% transparent. It's a lot harder the perpetrate white collar crime on it since every single transaction is public. No, it doesn't remove bad actors from the system, but it opens everything up for everyone to see, which helps.

Other than that I'm not really sure, tbh. Our centralized banking system seems way overdue for an upgrade though, don't you think? Payments and transactions shouldn't take days to clear (or longer). Why does nothing ever happen on weekends? Or at night? It should all be pretty much instantaneous and available to us 24/7. Isn't this supposed to be the information age? The banks live in the past.

Let me also note that if banks and other financial institutions didn't think they could use blockchain tech to improve their services, they wouldn't be buying in and incorporating the tech the way they are right now. As with any technology, there are likely improvements there that aren't obvious yet.

You could say that the blockchain is nothing more than a decentralized ledger. You could also say that electricity is nothing more than moving electrons though. The implications of these technologies require a lot more digging and research than just a cursory glance at the surface.



I mean, it is being used for all sorts of things right now, including, for example, helping companies clean up our oceans. Here is one example. There are more.

A programmable decentralized ledger can be used for a lot more than just finance related projects. And it is.

These problems with banking are entirely down to the banks not wanting to spend time and effort making them better. If the US government mandated that credit card payments in US banks have to be same day, the banks would make them same day. The technology has been there for decades. Blockchain doesn't change anything. Hell, if your complaints are about the systems being slow and inefficient, why would you want to replace them with another slow, inefficient system? Compared to conventional databases, anything blockchain based is always going to be slow, there's just simply too many extra steps that it has to go through. Same with things like a lack of transparency. That's a choice, not something inherent to non-blockchain methods. (And frankly, I think it's a pretty reasonable choice when it comes to managing money)

The only use case people have found so far the blockchain works better at than conventional databases is cryptocurrency. And given that the only major practical use people have found for cryptocurrency is for financial speculation, that's not exactly something to be proud of. For example, nothing about that "seachaintoken" you linked to has to be blockchain based in order to clean up the sea - the only thing being on the blockchain gains them is maybe a bunch of cryptobros blindly throwing money and hype.

Maybe someday this will change. Maybe someone will find a genuine use case for blockchain that isn't outclassed by other methods. And I'm not saying people shouldn't try to find such a case. But right now, it does nothing useful.

 

[Samson]

These problems with banking are entirely down to the banks not wanting to spend time and effort making them better. If the US government mandated that credit card payments in US banks have to be same day, the banks would make them same day. The technology has been there for decades. Blockchain doesn't change anything. Hell, if your complaints are about the systems being slow and inefficient, why would you want to replace them with another slow, inefficient system? Compared to conventional databases, anything blockchain based is always going to be slow, there's just simply too many extra steps that it has to go through.

From the point of view of the user we only have the choices that are presented to us. To do internet payments we can use the conventional banking system, which according to this costs ~5% (min est. 2.75%, max est. 6.85%) of the transaction value and comes with significant setup costs for both parties.

Or we can use crypto. On my last monero transaction I paid 0.0011% in fees and both parties just need a bit of software.

Which is the inefficient one?

 

[PhroX]

From the point of view of the user we only have the choices that are presented to us. To do internet payments we can use the conventional banking system, which according to this costs ~5% (min est. 2.75%, max est. 6.85%) of the transaction value and comes with significant setup costs for both parties.

Or we can use crypto. On my last monero transaction I paid 0.0011% in fees and both parties just need a bit of software.

Which is the inefficient one?

That's not the point I was making. The inefficiencies and costs in the banking systems are there because the banks either want them, or don't want to put time and money into changing them. They're not inherent to the underlying technology being used. Blockchain as a technology is less efficient that other forms of database. The issue is, basically, the usual one: greed. (And frankly, I can't see that greed disappearing if you switch to the blockchain. You put all transfers on the blockchain, and they put more fees on the blockchain.).

 

[Samson]

That's not the point I was making. The inefficiencies and costs in the banking systems are there because the banks either want them, or don't want to put time and money into changing them. They're not inherent to the underlying technology being used. Blockchain as a technology is less efficient that other forms of database. The issue is, basically, the usual one: greed. (And frankly, I can't see that greed disappearing if you switch to the blockchain. You put all transfers on the blockchain, and they put more fees on the blockchain.).

It is getting a bit philosophical, but when the "technology" is "having an trusted intermediary who keeps everyone's accounts" and the "inefficiencies and costs" are those that make the intermediary the most money, then it is reasonable to say the the costs are inherent in the tech.

Obviously a distributed ledger will always use more total electricity than a single relational database. If the distributed ledger can run on the users machines, and the single relational database requires a system than costs one in every 20 dollars it processes then the distributed ledger is more efficient for any meaningful definition of efficiency.

Sure it is greed that drives capitalism. You and I should be greedy with our money and not want to give more than we need to others. It is in out greedy self interest to transfer value the way that means we get to keep most of it. That means using the medium that costs less, and frequently than means crypto nowadays. We should not care about the greed of intermediaries we do not need.

 

[MrCynical]

From the point of view of the user we only have the choices that are presented to us. To do internet payments we can use the conventional banking system, which according to this costs ~5% (min est. 2.75%, max est. 6.85%) of the transaction value and comes with significant setup costs for both parties.

Or we can use crypto. On my last monero transaction I paid 0.0011% in fees and both parties just need a bit of software.

Which is the inefficient one?

You're mixing up two different things here:

1) The inherent cost of performing the transaction (electricity, computational power etc).

2) The fee charged for performing it (which will be larger in order to include a profit margin).

Bitcoin is infamously terrible on the first issue, consuming the electricity output of a decent size country for a total number of transactions which is a rounding error compared to those performed by conventional banking services. Other cryptocurrencies have tried to address this in various ways, and some are not so absurdly inefficient. However the concept of a blockchain containing a ledger of all transactions ever performed with that currency is fundamentally problematic when it comes to scaling up.

As for the second issue, sure banks charge these fees, but the numbers above are not the inherent costs of carrying out the transaction. This is self-evident from the fact they are percentages. The actual costs of processing a transaction are largely independent of the size of the payment.

Obviously a distributed ledger will always use more total electricity than a single relational database. If the distributed ledger can run on the users machines, and the single relational database requires a system than costs one in every 20 dollars it processes then the distributed ledger is more efficient for any meaningful definition of efficiency.

But as above, it doesn't require a system that costs 1 in 20 dollars. The actual processing costs are a tiny fraction of that (and is always going to be less than that of any blockchain, simply due to the amount of information required to be processed). The overwhelming majority of that cost is the bank's cut, and is at the level they can get away with charging and still have people use the system.

 

[Samson]

You're mixing up two different things here:

1) The inherent cost of performing the transaction (electricity, computational power etc).

2) The fee charged for performing it (which will be larger in order to include a profit margin).

Bitcoin is infamously terrible on the first issue, consuming the electricity output of a decent size country for a total number of transactions which is a rounding error compared to those performed by conventional banking services. Other cryptocurrencies have tried to address this in various ways, and some are not so absurdly inefficient. However the concept of a blockchain containing a ledger of all transactions ever performed with that currency is fundamentally problematic when it comes to scaling up.

As for the second issue, sure banks charge these fees, but the numbers above are not the inherent costs of carrying out the transaction. This is self-evident from the fact they are percentages. The actual costs of processing a transaction are largely independent of the size of the payment.



But as above, it doesn't require a system that costs 1 in 20 dollars. The actual processing costs are a tiny fraction of that (and is always going to be less than that of any blockchain, simply due to the amount of information required to be processed). The overwhelming majority of that cost is the bank's cut, and is at the level they can get away with charging and still have people use the system.

I am not mixing these things together, they are mixed together by the system. It is of no practical importance to the user that the actual computing system of the banks uses little electricity. It could use as much as the bitcoin network, and the user would hardly notice because the fees are so far beyond that. And as you say bitcoin is the absolute worst that I have not been suggesting people use as a medium of exchange. You cannot complain about other systems just because that is bad.

 

[MrCynical]

I am not mixing these things together, they are mixed together by the system. It is of no practical importance to the user that the actual computing system of the banks uses little electricity.

The practical point is that if the market does start to object to those fees, the banks have a lot of scope to cut them and still turn a profit. More scope than crypto does due to the actual processing cost being lower.

It could use as much as the bitcoin network, and the user would hardly notice because the fees are so far beyond that.

If all banking transactions used as much electricity as the bitcoin network, they would exceed the generating capacity of the entire world by a huge amount. I think users would end up noticing that one way or another.

 

[EnglishEdward]

Indeed there is a strong argument for making certain types of digital currency illegal because of their contribution to climate change.

 

[Samson]

The practical point is that if the market does start to object to those fees, the banks have a lot of scope to cut them and still turn a profit. More scope than crypto does due to the actual processing cost being lower.

In the current environment, what does "start to object to those fees" look like other than building and using decentralised payment solutions?

If all banking transactions used as much electricity as the bitcoin network, they would exceed the generating capacity of the entire world by a huge amount. I think users would end up noticing that one way or another.

I am talking specifically about remote sales transactions. I cannot make such a payment without the card companies taking about 5%. If those transactions had the electricity per dollar electricity consumption of bitcoin that would be a small amount of 5%. This just highlights how expensive card service are, not that bitcoin is a good payment solution.

 

[Moriarte]

Ah yes, smart contracts. Another example that bad ideas are disproportionately likely to have the word "smart" in the name. I have just enough experience coding, and having code behave in ways other than I intended, to find it horrifyingly hilarious when people say "the code is the law" like that's a good thing...

They are just automated contracts. I don’t immediately see why they’re such a bad idea, perhaps you can expand. In my bank I’ve been using an analogue to smart contract to pay for my cell phone, for a decade. Quite helpful.

But as above, it doesn't require a system that costs 1 in 20 dollars. The actual processing costs are a tiny fraction of that (and is always going to be less than that of any blockchain, simply due to the amount of information required to be processed). The overwhelming majority of that cost is the bank's cut, and is at the level they can get away with charging and still have people use the system.

Am I understanding correctly: we are excluding the lion’s share of the cost of maintaining worldwide established financial network (including bailouts; the cost and pollution by tens of millions within finance going to work) in favour of a theory, which says that a theoretical centralised computer is more efficient than a theoretical decentralised one?

Banks can’t cut costs, interbank competition has driven those costs so high, there’s barely margin to be made in the financial sector.

 

[Samson]

Banks can’t cut costs, interbank competition has driven those costs so high, there’s barely margin to be made in the financial sector.

Fancy explaining this? That is not usual economic theory.

 

[Moriarte]

Fancy explaining this? That is not usual economic theory.

Retail banks are established businesses, burdened by infrastructure, forced to compete with each other for clients, just like any other business. Were they able to cut costs and thus offer better transaction rates, they’d do it to get more clients.

 

[MrCynical]

In the current environment, what does "start to object to those fees" look like other than building and using decentralised payment solutions?

It would look like some significant percentage of all retail transactions moving to an alternative system with lower fees. A system whose primary use was to process such transactions. No cryptocurrency currently fills that role. Firstly the fractions of currency transactions using crypto are miniscule compared to the economy as a whole. Secondly, no-one is creating and using cryptosystems with the idea of such fees being the source of profit. The expectation for making money from crypto is still as speculative tokens, with the idea that one will "go to the moon!" The vast majority of cryptocurrency transactions are still not using it as a currency.

The cynic in me also points out that the path of least resistance, even if a cryptocurrency starts functioning as a currency, is for the transaction fees to rise to be competitive - a bit lower perhaps, but not spectacularly lower - than existing banking methods.

Retail banks are established businesses, burdened by infrastructure, forced to compete with each other for clients, just like any other business. Were they able to cut costs and thus offer better transaction rates, they’d do it to get more clients.

Like Samson, I'm scratching my head a bit, as your last comments seem to be arguing that intense competition is driving prices up. Which is kind of a direct contradiction of typical economic ideas. It would make a lot more sense if you were arguing banks function as a cartel - actively or passively - to keep their rates high.

Am I understanding correctly: we are excluding the lion’s share of the cost of maintaining worldwide established financial network (including bailouts; the cost and pollution by tens of millions within finance going to work) in favour of a theory, which says that a theoretical centralised computer is more efficient than a theoretical decentralised one?

You're imagining you won't get this kind of financial network and bailouts if crypto is running even a substantial minority of retail transactions? Recent collapses of exchanges already showed it's pretty much the same game as conventional banking on that score. It's just that there's less incentive to bailout crypto at present, as it's a niche market, and one with a public perception that it's your own stupid fault if you lose all your money by getting involved in it. The "let it burn" approach won't be an option if basic retail is relying on it.

As for centralised vs decentralised computing, you're missing the point that the decentralised computer is having to process vastly more information per transaction (the blockchain ledger). And that the amount of information increases proportionally the more transactions are processed. Ideas of blockchain being even theoretically sustainable rely on the idea that computer technology is improving fast enough to keep up with the constant increase in processing required, which seems very dodgy to me.

 

[Samson]

It would look like some significant percentage of all retail transactions moving to an alternative system with lower fees. A system whose primary use was to process such transactions. No cryptocurrency currently fills that role. Firstly the fractions of currency transactions using crypto are miniscule compared to the economy as a whole. Secondly, no-one is creating and using cryptosystems with the idea of such fees being the source of profit. The expectation for making money from crypto is still as speculative tokens, with the idea that one will "go to the moon!" The vast majority of cryptocurrency transactions are still not using it as a currency.

The tech is not there to fully replace cards yet, but how long did it take cards to mostly replace physical money?

I am not arguing that we should today all switch to monero for everything. I am saying that transfering money by cards is vastly less efficient than transfering money by monero, for a meaningful definition of efficiency.

The cynic in me also points out that the path of least resistance, even if a cryptocurrency starts functioning as a currency, is for the transaction fees to rise to be competitive - a bit lower perhaps, but not spectacularly lower - than existing banking methods.

Why would this happen, if the change is driven by individuals writing and using open source software? It is obvious why it happens in a centralised system, but that is exactly why we should have an alternative.

Is it not more likely that the centralised systems will be forced to become more efficient? If that is really the way to go, could not fees fall to lower than decentralised solutions? Wouldn't that be better for everyone?

 

[Moriarte]

Like Samson, I'm scratching my head a bit, as your last comments seem to be arguing that intense competition is driving prices up.

Prices of what? Intense competition drives profit margins down. That’s why there is no scope for substantial transaction cost cuts in traditional finance. Crypto can be centralised or decentralised, but it is initially unburdened by the need to hold vast apparatus. So, you know, in 50 years it can become a thing

It would make a lot more sense if you were arguing banks function as a cartel - actively or passively - to keep their rates high.

That is a separate issue. Indeed they do. Perhaps not globally, but regionally. On a global scale, banking establishments still have to compete.