Cryptocurrency Mining Consumes More Power Than 17M Population Country

[R-T-B]

Not likely.

Dude. Just google "bank blockchain tech." You could not be more wrong.

 

[cdawall]

Bitcoin is legal in many countries, but it's not a legal tender. I'm also including the wiki link (like @FordGT90Concept did) in case you get bored with flashing mining BIOS etc.

Cough

https://totalbitcoin.org/bitcoin-legal-tender-japan/

Cough

 

[notb]

Cough
https://totalbitcoin.org/bitcoin-legal-tender-japan/
Cough

You said "many"... but whatever. Let's talk about Japan.

I have to admit that I didn't have enough time to read the actual document (is it even available in English?). I only went through some interpretations and they didn't say that BTC becomes a legal tender, but just a legal methods of payment. This is a totally different thing. But I'll have to check that to be sure.

Other things described in this article are in line with what I've read earlier and... to be honest, I don't fully understand why are you so happy about this act. It's against everything you seem to like about BTC.
1) This act creates something called a "virtual currency", which is neither a real currency nor an electronic payment system. In fact it inherits most properties from commodities, so it's a de facto commodity - like in most other countries already.
2) Where Japan's act differs from those in other countries, is that it is very strict.
It requires bitcoin-trading organizations to fulfill the same requirements that commodity exchanges have to. This includes regular reporting, auditing, robust IT systems, implementing many regulations - generally speaking a proper corporate structure that e.g. a bank has. It also means that the exchange has to store and - if asked - report to the authority all information on transactions (also naming the parties!).
Before this, a few IT guys were able to run a BTC exchange. Now this exchange will have to employ a 100 people just to make it legal. Will this still be profitable? Will anyone - apart from large financial institutions - be able to run a BTC exchange?
3) The act also changes the requirements for traders - that's you. You can still mine cryptocurrencies freely, but to exchange them for money in Japan you have to legalize your activity, i.e. sign a contract with a broker.

 

[FordGT90Concept]

Thing is though, these transactions between banks are organized not in real time, but always in sessions (there are just some workarounds that can simulate real-time confirmation).

To make double spending an impossibility. ACH holds the funds until both parties agree to the transfer, then it clears.

Checks are instructions to banks to exchange funds on your behalf.

Dude. Just google "bank blockchain tech." You could not be more wrong.

Reasons not to blockchain:
-computational cost is astronomical/inefficient
-nodes can still exploit the network fabricating debts for illgotten profit
-there's no authority to fix a problem. Case in point, if a vulnerability is discovered in Bitcoin's hashing algorithms (e.g. a collision), the entire chain will come crashing down.
-Bitcoin only handles maybe a quarter million transactions per day: a country like the USA has billions. Centralization means conservation of resources--no more redundancy in the system than is necessary.
-Looking at the statistics, it almost appears that doing more than a quarter million transactions per day is impossible because of the amount of overhead required to accomplish it.

TL;DR: Blockchains only work in confined situations where distributing a database is preferable to maintaining one--where the costs of creating an authority outweigh the needs of the project. If the project grows too much, blockchains become a liability.


Edit: ~$25/transaction right now. That's very expensive.

Edit: Similar list (and far more thorough) to what I compiled above: https://letstalkpayments.com/the-cost-of-saving-the-world-with-blockchain/

 

[R-T-B]

Reasons not to blockchain

You proceed to list a bunch of reasons that are not applicable to a small scale node set like banks are considering.

nodes can still exploit the network fabricating debts for illgotten profit

Uh no. Not without a 51% attack anyways. I'm sure banks are looking at 8 or so geographically seperated nodes to prevent this.

there's no authority to fix a problem. Case in point, if a vulnerability is discovered in Bitcoin's hashing algorithms (e.g. a collision), the entire chain will come crashing down.

Wrong, even with the current non-banking environment (and certainly in a banking environment). The "authority" is the software update process, and already bitcoin has demonstrated this works with a few hardforks.

Bitcoin only handles maybe a quarter million transactions per day: a country like the USA has billions.

Raising the block size could fix this easily (Bitcoin is about to do this) and has nothing restricting you from doing both that AND keeping node count low. Blockchain is more inefficient than centralization, but not so much as you assume. I have little / no doubt you could run it at no more than 2-4x the power cost of a credit card network, and gain a lot of resiliency to hacking and single point of failure problems in the process, making it VERY worth it.

 

[notb]

To make double spending an impossibility. ACH holds the funds until both parties agree to the transfer, then it clears.

Checks are instructions to banks to do exchange funds on your behalf.

Exactly. And most of the world haven't used them for years. This is quite weird that Americans still do. But then again: magnetic stripe cards, slow adoption of contactless payments, no instant electronic transfer... you're very traditional.
Many countries in the world are already preparing for cashless reality.
I'm not surprised at all that Japan is the first country with solid bitcoin regulation. They were always leading the electronic payment innovation and implementation.

Reasons not to blockchain:
-computational cost is astronomical

It doesn't have to be. Current cryptocurrencies are designed this way, but it's not a must.

-nodes can still exploit the network fabricating debts for illgotten profit

What exactly do you mean?

-there's no authority to fix a problem. Case in point, if a vulnerability is discovered in Bitcoin's hashing algorithms, the entire chain will come crashing down.

Again, this is stemming from how bitcoin is designed. A closed, bank-run blockchain doesn't have such issues. Only the bank will do hashing.

-Bitcoin only handles maybe a quarter million transactions per day: a country like the USA has billions which translates into astronomical costs. Centralization means conservation of resources--no more redundancy in the system than is necessary.
-Looking at the statistics, it almost appears that doing more than a quarter million transactions per day is impossible because of the amount of overhead required to accomplish it.

But you're talking about issues of Bitcoin, not blockchain in general.
This can be all addressed by a more specialized, well-designed blockchain solution.

Also keep in mind that even without blockchain the financial system is very demanding computation-wise. Blockchain is not going to be an addition. It will replace some currentlu used technologies.
On the other hand, we have to remember, frankly, banking technology hasn't evolved much during the last decade.
Compare today to the pre-crisis times in 2007. There are not many more humans (+15%) and they have similar financial needs. The number of traded stocks and other financial instruments didn't grow either. The instruments are not getting more complicated and 10 years ago we already had things like high speed algorithmic trading.
So what has changed? The computers did: they're 20-30 times faster.

Also don't worry about profitability of this idea. People mining at home are often not rational. They can't forecast properly, they can't estimate risk - financial institutions can.

 

[FordGT90Concept]

Uh no. Not without a 51% attack anyways. I'm sure banks are looking at 8 or so geographically seperated nodes to prevent this.

Every organization that processes a significant number of transactions has to have it's own node. The 2007 financial collapse was orchestrated by most of the big private banks. With the inability to reasonably regulate it, I expect massive and severe exploitation of the system.

Wrong, even with the current non-banking environment (and certainly in a banking environment). The "authority" is the software update process, and already bitcoin has demonstrated this works with a few hardforks.

And if said software update has a bug or a fake authority pushes the update out? NSA used fake security certificates to infiltrate Iran, for example.

Raising the block size could fix this easily (Bitcoin is about to do this) and has nothing restricting you from doing both that AND keeping node count low. Blockchain is more inefficient than centralization, but not so much as you assume. I have little / no doubt you could run it at no more than 2-4x the power cost of a credit card network, and gain a lot of resiliency to hacking and single point of failure problems in the process, making it VERY worth it.

Increasing network and processing load which again leads to costs running out of control.

Exactly. And most of the world haven't used them for years. This is quite weird that Americans still do. But then again: magnetic stripe cards, slow adoption of contactless payments, no instant electronic transfer... you're very traditional.
Many countries in the world are already preparing for cashless reality.
I'm not surprised at all that Japan is the first country with solid bitcoin regulation. They were always leading the electronic payment innovation and implementation.

Actually, most of the world's financial systems work on clearing houses but they're hidden. The delays are intentional: they protect buyers and sellers. Credit cards have transactions cleared within minutes because it's literally a bank-to-bank transfer. They are the exception because the banks themselves enforce fraud protection.

Most of US transactions do occur cashlessly.

It doesn't have to be. Current cryptocurrencies are designed this way, but it's not a must.

It does because it's peer-to-peer. Cryptography is what creates the "trust" mechanisms.

What exactly do you mean?

51% attack. Fool the network into believing your data is legitimate and it has no way of knowing differently.

Again, this is stemming from how bitcoin is designed. A closed, bank-run blockchain doesn't have such issues. Only the bank will do hashing.

Again, peer-to-peer. If banks are going to be doing it exclusively then there's absolutely no reason to blockchain in the first place. Blockchains are just decentralized databases.

But you're talking about issues of Bitcoin, not blockchain in general.
This can be all addressed by a more specialized, well-designed blockchain solution.

No, cryptocurrencies add a crapload of computational and network overhead compared to traditional databases. There's no getting around that fact.

Also keep in mind that even without blockchain the financial system is very demanding computation-wise.

Never took accounting classes? Financial transaction systems are very foundationally simple. Right now, it really boils down to five data points: source routing number, source account number, destination routing number, destination account number, amount. At each bank, it's a simple floating point addition or subtraction. This is why it is not difficult for banks to process billions of these transactions in a day. It really requires very little in terms of hardware. There's also hardware out there specifically designed to process these kinds of transactions in databases (e.g. SPARC).

 

[notb]

Actually, most of the world's financial systems work on clearing houses but they're hidden. The delays are intentional: they protect buyers and sellers. Credit cards have transactions cleared within minutes because it's literally a bank-to-bank transfer. They are the exception because the banks themselves enforce fraud protection.

Sadly. no. Credit card transactions are also cleared during sessions - usually once a day. You only get the authentication quickly and your bank locks the payment amount on your account.

It does because it's peer-to-peer. Cryptography is what creates the "trust" mechanisms.

And banks do the hashing, but this won't have to be so computation-heavy since there will be not many parties in the system.

Again, peer-to-peer. If banks are going to be doing it exclusively then there's absolutely no reason to blockchain in the first place. Blockchains are just decentralized databases.

This is exactly the reason why they want to do this. A decentralized database is fast and transparent. It will make instant transfers between banks possible.

I think you're generally thinking too much about financial applications - they're fairly difficult to understand.

There are many initiatives to use blockchain in logistics. Example:
http://www-03.ibm.com/press/us/en/pressrelease/51712.wss

Never took accounting classes? Financial transaction systems are very foundationally simple.

I assure you it's not that simple. There's a lot more to accounting systems than just tracking how much money is on a particular account.
And of course I didn't take any accounting classes. Accounting is pretty boring. :-D

 

[R-T-B]

And if said software update has a bug or a fake authority pushes the update out? NSA used fake security certificates to infiltrate Iran, for example.

And how is this applicable to a closed banking environment, or have any relevance to the original topic, which was the bogus claim that there is "no authority to fix things?"

No, cryptocurrencies add a crapload of computational and network overhead compared to traditional databases. There's no getting around that fact.

Not as much as you think. The fact that big name banks are doing blockchain tests should tell you how far off you are in this assumption.

51% attack. Fool the network into believing your data is legitimate and it has no way of knowing differently.

Uh... do you even know what a 51% attack is? It means you control over 50% of the nodes, literally. In an 8 machine network you'd need to compromise 5 seperate machines. This is a BENEFIT, not a problem.

From what I'm reading from you Ford, you need to seriously advance your understanding of how this works before debating further. No offense meant. You raised some good philosophical issues of which I even conceeded to a few, but your technical understanding of blockchain appears more driven by bias than knowledge.

In particular, I suggest you study proof of work and diffilculty, which is where the vast exponential majority of energy goes (and ironically, has almost nothing to do with validating transactions, and so could be removed).

EDIT: Or are you assuming ford someone breaches the mining network and floods it with hashrate to 51% it? Unlikely, given it is probably IPSEC'd if not airgapped.

 

[FordGT90Concept]

Sadly. no. Credit card transactions are also cleared during sessions - usually once a day. You only get the authentication quickly and your bank locks the payment amount on your account.

Actually, it is much slower than a day for the transaction to completely finish. The credit account gets a hold for the amount of the transaction within minutes which, as far as the consumer is concerned, the transaction is complete (purchase accessible immediately). It still takes days for the hold to vanish and the funds to transfer to the recipient but the purchaser has no way of knowing that without logging into the creditor's website and looking at the pending transaction. The bank is good for the money which is why retailers don't feel compelled to wait until the transaction finishes.

Checks work the same way (they have faith in you that the check won't bounce). If it bounces, they can use the information on the check to get what their owed.

This is exactly the reason why they want to do this. A decentralized database is fast and transparent. It will make instant transfers between banks possible.

Consumers have a right to privacy. Decentralized databases are not fast. Near instant transfers can already be accomplished via ACHs but we choose not to for buyer and seller protection.

And how is this applicable to a closed banking environment, or have any relevance to the original topic, which was the bogus claim that there is "no authority to fix things?"

You're pushing for a software solution to a problem that addressed by many moving parts (namely, there's a lot of people involved in the financial industry).

Not as much as you think. The fact that big name banks are doing blockchain tests should tell you how far off you are in this assumption.

Banks are always looking for ways to increase profit (cut costs). Of course they're going to take a look at it. Doesn't mean it will go beyond prototyping or be confined to minor networks like ATMs.

Uh... do you even know what a 51% attack is? It means you control over 50% of the nodes, literally. In an 8 machine network you'd need to compromise 5 seperate machines. This is a BENEFIT, not a problem.

...or the software that runs on all of them. Or the hardware. Or the kernel. The smaller the network, the more vulnerable it is. Still makes more sense to have eight traditional databases that update each other. You know, like the fed already does.


I'm getting the distinct impression neither of you looked at that link I gave you. TL;DR: blockchain will not be used for financial transactions in the USA any time soon, if ever.

 

[R-T-B]

You're pushing for a software solution to a problem that addressed by many moving parts (namely, there's a lot of people involved in the financial industry).

I'm not sure I'm pushing for anything. I believe in some things future, perhaps, but I don't neccesarily have a stake in it or care either way. I don't see a need if the banks don't. I just want people to have their facts straight before they jump into anything.

Banks are always looking for ways to increase profit (cut costs). Of course they're going to take a look at it. Doesn't mean it will go beyond prototyping or be confined to minor networks like ATMs.

But it's inherently inefficient, remember? Or is it?

...or the software that runs on all of them. You know why USA's power grid is durable? Because it is diverse. No one attack is going to have very far reaching impact. There isn't one digital key that controls the whole system. The US banking system is the same. Relying on digital system to handle all of a country's *anything* is foolish.

Well, yeah. But do you really think if a closed blockchain were adopted, all the various banks would not do their own closed systems? Can you really see Chase and Bank of America sharing books?

Oh, and the power grid... heh:

https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

Not entirely relevant, but good reading.

 

[FordGT90Concept]

But it's inherently inefficient, remember? Or is it?

ATMs are often remote, not very numerous, and have very few transactions on them in an average day. Problem is communicating with the main branch in a way that prohibits double spending and without giving any more information to the chain than is necessary. Probably why they're prototyping.

Well, yeah. But do you really think if a closed blockchain were adopted, all the various banks would not do their own closed systems? Can you really see Chase and Bank of America sharing books?

I expect most won't leave the systems they're currently on. The cost is too great and the benefits are virtually nonexistent.

Oh, and the power grid... heh:

https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

Not entirely relevant, but good reading.

"256 power plants are off-line, 85% of which went offline after the grid separations occurred, most due to the action of automatic protective controls." It was the grid responding to the grid to prevent damage.

 

[R-T-B]

"256 power plants are off-line, 85% of which went offline after the grid separations occurred, most due to the action of automatic protective controls." It was the grid responding to the grid to prevent damage.

Oh, I know. I just can't hear of "grid resiliency" without thinking of that. I did warn you it wasn't really relevant.

The cost is too great and the benefits are virtually nonexistent.

That's where I think your lack of knowledge is really hurting you, as I don't believe that to be the case from any of the basic principles, but I suppose only time will tell.

 

[notb]

Actually, it is much slower than a day for the transaction to completely finish. The credit account gets a hold for the amount of the transaction within minutes which, as far as the consumer is concerned, the transaction is complete (purchase accessible immediately). It still takes days for the hold to vanish and the funds to transfer to the recipient but the purchaser has no way of knowing that without logging into the creditor's website and looking at the pending transaction. The bank is good for the money which is why retailers don't feel compelled to wait until the transaction finishes.

I don't think the consumer's point of view is relevant here. Blockchain solutions, at least at first, will be used in internal operations and inter-bank transfers.

Consumers have a right to privacy. Decentralized databases are not fast. Near instant transfers can already be accomplished via ACHs but we choose not to for buyer and seller protection.

You're mixing things, so I already don't know whether you mean the actual transfer (important for banks and sellers) or just the authorization (important for the client).

And I don't understand why you mention privacy. Blockchain (distributed database) won't change anything from how this works currently (i.e. on local databases).

Banks are always looking for ways to increase profit (cut costs). Of course they're going to take a look at it. Doesn't mean it will go beyond prototyping or be confined to minor networks like ATMs.

It's already being implemented in industry (it can be used in basically any system where something is moving between parties). It will take longer for banks because of all the regulation, but it's almost inevitable.

 

[FordGT90Concept]

You're mixing things, so I already don't know whether you mean the actual transfer (important for banks and sellers) or just the authorization (important for the client).

Both. Because all of the systems are already electronic (banks and ACH), ACH could completely clear transactions in less than a second. More light reading: https://www.depositaccounts.com/blog/inside-look-at-ach-transfer-speeds-at.html
TL;DR: banks are causing the delays, not ACH itself.

And I don't understand why you mention privacy. Blockchain (distributed database) won't change anything from how this works currently (i.e. on local databases).

The local databases aren't transparent. They're behind lock and key only subject to view of a very limited number of people.

 

[notb]

Both. Because all of the systems are already electronic (banks and ACH), ACH could completely clear transactions in less than a second. More light reading: https://www.depositaccounts.com/blog/inside-look-at-ach-transfer-speeds-at.html
TL;DR: banks are causing the delays, not ACH itself.

You keep saying that ACH could clear transactions in seconds, but you fail to give any proof. :-D
This article states that 1-2 days are considered fast, 3-4 is slow.

I have to say: the times mentioned here are just sad. A transfer between different countries in Europe used to take 2 days for a while now. A transfer to the same country is under 1 day - even few hours if you do it in the morning.
Later this year a new system for transfers within EU is expected to launch (SEPA, covering few non-EU countries as well). It will offer instant transfers (10 seconds), but full clearance will still take a bit more. Similar systems already exist in particular countries. In Poland an "express transfer" takes under 15 minutes and is works 24/7 (unlike Mon-Fri for normal transfers). But there is a premium for this, while normal "1-day" transfers usually are free.

You can easily identify fast transfer systems that don't do full clearance within the transfer time: they have low upper limits for the payment amount.

Before state-organized fast transfers were launched in Poland, there was even a private company offering such a product. How? Intra-bank transfers are instant, so it had an account in each major bank with a significant sum of money (~$250k) and money went through them. It wasn't perfect, but still a very nice concept.

The local databases aren't transparent. They're behind lock and key only subject to view of a very limited number of people.

And so will be the blockchain-based system. What exactly don't you understand?

 

[FordGT90Concept]

This article states that 1-2 days are considered fast, 3-4 is slow.

Because the sending bank likes to hold not to the money as long as they can before sending it. Initiating the transfer as a credit is much faster. If banks aren't delaying, ACH normally clears transactions in less than a day.

I have to say: the times mentioned here are just sad. A transfer between different countries in Europe used to take 2 days for a while now. A transfer to the same country is under 1 day - even few hours if you do it in the morning.
Later this year a new system for transfers within EU is expected to launch (SEPA, covering few non-EU countries as well). It will offer instant transfers (10 seconds), but full clearance will still take a bit more. Similar systems already exist in particular countries. In Poland an "express transfer" takes under 15 minutes and is works 24/7 (unlike Mon-Fri for normal transfers). But there is a premium for this, while normal "1-day" transfers usually are free.

ACH is not supposed to take more than 3 days. If it does, you're supposed to complain to the bank that initiated the transfer.

And so will be the blockchain-based system. What exactly don't you understand?

The entire database rarely, if ever, leaves the facility with local databases. Blockchain, the entire database is constantly flying around the internet.

 

[Fx]

Are people really making a profit after paying costly monthly power bills?

 

[FordGT90Concept]

Presumably yes otherwise they wouldn't be doing it. The cards cost more than the electricity in most places that are doing it.

 

[R-T-B]

The entire database rarely, if ever, leaves the facility with local databases. Blockchain, the entire database is constantly flying around the internet.

What part of "closed system" (which is what the banks are considering) don't you understand? Blockchain can just as easily be run on a closed lan.

 

[FordGT90Concept]

If it is closed, blockchain has no advantages over a traditional database.

 

[notb]

Because the sending bank likes to hold not to the money as long as they can before sending it. Initiating the transfer as a credit is much faster. If banks aren't delaying, ACH normally clears transactions in less than a day.

This just shows that banks don't cooperate very well in US or they really don't care about customers. This is a zero-sum game. Quick transfers would not hurt their balance.
Or this could be because of cheques, i.e. because they take so long to confirm, banks slow down the electronic transfers as well. Otherwise a bank that sends transfers "physically" would have an advantage.
Once again: in Europe transfers used to take a day, often less. It's most likely similar in other developed areas.

ACH is not supposed to take more than 3 days. If it does, you're supposed to complain to the bank that initiated the transfer.

A side note: it does look a bit like if you were advocating ACH in this discussion - even though the whole transfer system in US clearly seems fairly outdated and ineffective.. Why is that? You've written some of the code or what?
Because you're quite happy with bashing banks, so it's clearly not a general fondness for the financial sector.

The entire database rarely, if ever, leaves the facility with local databases. Blockchain, the entire database is constantly flying around the internet.

It's not the whole database from banks' point of view. It's just the part about transactions. It can't, for example, allow an identification of a client - this will only be possible in the bank that operates this client's account.

Presumably yes otherwise they wouldn't be doing it. The cards cost more than the electricity in most places that are doing it.

Most likely not. My impression is that a lot of home-miners can't calculate the expected profit properly. They, so to speak, go with the flow.

If it is closed, blockchain has no advantages over a traditional database.

Of course it has: speed, transparency, potentially lower maintenance cost - things like that hold. They're even improved by closing the blockchain, since it can use simpler hashing algorithms.

 

[FordGT90Concept]

This just shows that banks don't cooperate very well in US or they really don't care about customers. This is a zero-sum game. Quick transfers would not hurt their balance.
Or this could be because of cheques, i.e. because they take so long to confirm, banks slow down the electronic transfers as well. Otherwise a bank that sends transfers "physically" would have an advantage.
Once again: in Europe transfers used to take a day, often less. It's most likely similar in other developed areas.

A side note: it does look a bit like if you were advocating ACH in this discussion - even though the whole transfer system in US clearly seems fairly outdated and ineffective.. Why is that? You've written some of the code or what?
Because you're quite happy with bashing banks, so it's clearly not a general fondness for the financial sector.

As I said repeatedly, the delay is a window to void the transaction (payee, recipient, law enforcement, credit agency, bank, etc.).

ACH could clear transactions much faster but the delay is intentional.

It's not the whole database from banks' point of view. It's just the part about transactions. It can't, for example, allow an identification of a client - this will only be possible in the bank that operates this client's account.

...which every check points directly at (routing and account number). In other words, you hand a check to someone, you're also handing them everything they need to find every blockchain-recorded transaction you've ever made.

Of course it has: speed, transparency, potentially lower maintenance cost - things like that hold. They're even improved by closing the blockchain, since it can use simpler hashing algorithms.

Local databases are much faster than distributed databases. Transparency violates privacy. Maintenance is only lower cost if it is being farmed out to someone else (in which case, the cost is hidden and still not cheaper).

 

[notb]

As I said repeatedly, the delay is a window to void the transaction (payee, recipient, law enforcement, credit agency, bank, etc.).

ACH could clear transactions much faster but the delay is intentional.

Now this is pretty weird. Why so much control is needed for electronic payments?
I don't get your logic. Can you give an example of a situation where this becomes important?

...which every check points directly at (routing and account number). In other words, you hand a check to someone, you're also handing them everything they need to find every blockchain-recorded transaction you've ever made.

No, because my name is not in the blockchain. There are only some IDs that just the bank can translate.
I think you still misunderstand how this works. Bank-created blockchain will not be available to me to pay for, for example, Steam games. The thing we're talking about now is just for inter-bank transfers. It's replacing one technology with another.
If banks decide to make a blockchain for clients, it'll work just like bitcoin - with all its pros and cons - apart of one: it'll have a stable value (most likely a constant one).

Local databases are much faster than distributed databases. Transparency violates privacy. Maintenance is only lower cost if it is being farmed out to someone else (in which case, the cost is hidden and still not cheaper).

Maintenance will be lower because this will be a cooperation of banks, so the cost will spread. Also there will be no additional cost of the interface. Today banks pay the "transfer hubs" (like ACH in your case). With blockchain they won't have to.

BTW: just from a rational POV - if you're right and distributed ledger has no advantages and won't reduce cost, why are banks so interested?
Why is this happening? en.wikipedia.org/wiki/R3_(company)
Any theory?

 

[FordGT90Concept]

Now this is pretty weird. Why so much control is needed for electronic payments?
I don't get your logic. Can you give an example of a situation where this becomes important?

Example: someone steals your checkbook, writes a check to him/herself, and deposits it. A day later, you realize the checkbook is missing and notify your bank. The bank closes the account and gives you a new one. If ACH was instant, that money would be gone forever. Because ACH is not instant, you and your bank have time to prevent any damage from being done. The same applies to debit card transactions.

No, because my name is not in the blockchain. There are only some IDs that just the bank can translate.

Which are the same numbers the bank requires to access your account. Even if the account number is hidden behind a separate ID, that ID can be used to find all transactions someone made and figure out who they are. A public ledger is not an option. There can't be more transparency than the existing system.

If banks decide to make a blockchain for clients, it'll work just like bitcoin - with all its pros and cons - apart of one: it'll have a stable value (most likely a constant one).

It has to have null value. People don't currently pay, directly, for ACH or banking (unless you use something like American Express). Banks and credit unions pay the ACH/banking directly. Reason: they make a lot of money on loan interest.

Maintenance will be lower because this will be a cooperation of banks, so the cost will spread. Also there will be no additional cost of the interface. Today banks pay the "transfer hubs" (like ACH in your case). With blockchain they won't have to.

All that does is hide the costs, not reduce or remove them. And as I said repeatedly, ACHs aren't going away.

BTW: just from a rational POV - if you're right and distributed ledger has no advantages and won't reduce cost, why are banks so interested?
Why is this happening? en.wikipedia.org/wiki/R3_(company)
Any theory?

The link speaks loudly enough...

In November 2016, Goldman Sachs, Santander and Morgan Stanley each withdrew from the consortium.

Goldman Sachs was one of the first to join the consortium and one of the biggest banks in the USA. Pretty indicative it isn't a platform they're going to use.

Also in May 2017, the Bank of Canada and Payments Canada, as well of some Canadian chartered banks, announced the results of a year-long trial of R3 consortium's DLT (Distributed Ledger Technology) which will not be adopted at this time for reasons of transaction privacy and scalability.

Uh huh.

Corda is described as a distributed ledger, which "was never designed to be a traditional blockchain platform".

Yup.

TL;DR: Bank/credit union ledgers shouldn't be distributed.