5 ms·

The title is a bit misleading. This new "USC" (utility settlement coin) is for inter-bank transactions only, not for the general public. It apparently is back

The title is a bit misleading. This new "USC" (utility settlement coin) is for inter-bank transactions only, not for the general public. It apparently is backed by a blockchain as a distributed ledger. Details are a bit sparse across the web, but apparently it is intended as a way for fast money transfer between and within central banks and regular banks. The coins act as a proxy for 'real' money issued by central banks.

pzone 10y ago
Yes, this is extremely misleading. If a few banks want to replace their current transaction settlements with a blockchain protocol tailored to their needs, that's great, I hope it works out for them. Worrying that this blockchain isn't "open" or "democratic" like Bitcoin is just silly. It's like worrying about the fact the NYSE hasn't open sourced its internal code.
- dogma1138 10y ago
  Bitcoin isn't exactly "democratic" also, the size of the blockchain the the difficulty of the mining pretty much means that anyone who wants to get into the business would have to invest huge sums of capital to get into the game. The rest of us would have to use intermediate parties that would handle everything for us.
  - pzone 10y ago
    Right. I use my credit cards and simply trust Chase/Citi/etc to provide the services they've promised. I use a website to do Bitcoin transactions and simply trust they are not going to run away with my money. So there's even less of a justification to complain that this system is controlled only by powerful players.
  - repomies691 10y ago
    Hmm, currently the block size seems still manageable to me (80GB). Also you can turn on pruning so you need to store only the amount you want, and you can still use all the major use cases. You don't need to mine bitcoin to use it. It is still very cheap to use bitcoin still as an independent node. You can use those gateways to make your life a bit easier, but from cost perspective I think they are not needed.
    - dogma1138 10y ago
      These things are eventually going to exponentially increase, if Bitcoin gets to even 1% of the global payment cards daily transaction rate you aren't going to be running nodes on your own. Now you can still get into running a node and even doing mining however as time passes the barrier to entry increases exponentially and considerably faster than the capacity of individual hardware so you can't even really catch up.
      - wongarsu 10y ago
        >These things are eventually going to exponentially increase While bandwidth and storage costs exponentially decrease. Handling the entirety of VISA's volume with bitcoin transactions would require at least a 4Mbit/s connection (about double for peak times). That's not exactly out of reach of normal consumers. In that scenario the blockchain would grow bigger by 40Gb each day, or 15Tb per year. If you want to store that all, with current hard drive prices that would be $40 per month for new hard drives. Of course if you're not interested in looking up other people's random transactions, most of the data can be pruned as soon as the coins are spent elsewhere, giving you barely any blockchain size increase. And all of those resources get cheaper all the time. Resource requirements aren't a fundamental problem for bitcoin. The block size that limits it to ~0.3% of VISA's volume, and the mining system that favours a near-monopoly for the person with the lowest operating costs are far bigger problems.
      - louprado 10y ago
        "If Bitcoin gets to even 1% of the global payment cards daily" Visa alone processes between 2000 and 10000 transactions a second. Bitcoin is architected to handle 7 transactions per second. From my understanding this limitation can not be overcome by adding more miners, it is by design.
    - uncletammy 10y ago
      I think you mean "blockchain" rather than "block size".
  - galdosdi 10y ago
    As an aside, I think people use intermediate parties unnecessarily way too much with bitcoin. Yes, mining it yourself is not feasible so you do need one to get ahold of bitcoin and to get fiat currency back from bitcoin, but you don't need (and maybe shouldn't use) one just for storing bitcoin, paying with bitcoin, receiving bitcoin. That's the whole damned point of bitcoin, but then people go and store it with mtgox anyway even though they're not about to exchange it, and are bummed out when mtgox loses their money. It sucks, but it's funny that half the point of bitcoin is making this situation impossible, if you bother to do it right (Of course, if you trust a third party more than you trust your own ability to not lose/forget your private key, then I guess it makes sense, but you're trading a lot off in exchange for that)
  - 20yrs_no_equity 10y ago
    You don't need to use an intermediate party to use bitcoin. You can submit transactions directly to the network from your wallet software without running a miner. The miners secure the network, but the nature of the system is that they are trustless, so long as %51 of them are honest. By having the cost of mining being high, that makes the cost of an attack on the network high. Which aligns the miners incentives (Eg: keep the price of bitcoins from going to zero as it would if the network were compromised) aligned with yours. The blockchain is a bit large at 80GB but you don't need to download the whole blockchain to run a wallet and make transactions there are "light wallets" that will do this, and currently in testing is an even more efficient method of doing this called segregated witness.
dogma1138 10y ago
This doesn't exactly act like a proxy currency (there would be legal issues with that). Banks are really interested in using blockchain like technologies for settlement and clearinghouse related tasks, it would allow them to have near real time settlements without much effort and could both reduce their liability and costs. Both concurrent and correspondent banking has pretty big liabilities banks have to keep standing accounts with funds to facilitate, these funds are pretty expensive for the banks to keep since they both do not earn any interest and they are not fractional. Banks are also tied to fairly expensive transfer schemes (ACH/FSP) and service providers like SWIFT because a lot of the fast transfer schemes have limited membership (there are only 10 clearing members in the FSP for example) most banks end up paying pretty large transfer fees for agency banks that perform the transactions for them. Having the ability to set up blockchains that could easily be extended which would provide both the messaging/transfer functionality and the ledger that can be used for account settlement and consolidation between banks, as well as being extensible enough to allow multiple banks to join in either to the main blockchain or to sub-chain makes all the sense for them. Banks however would not and for the most could not use "digital currencies" for their day to day dealings, the only currency they are allowed to "create" is their respective national currency.
- eru 10y ago
  > [...] the only currency they are allowed to "create" is their respective national currency. Which jurisdictions are you talking about here?
  - dogma1138 10y ago
    If bank start creating their own currencies it is very likely to conflict with the fractional lending model used by virtually all modern banking systems.
    - eru 10y ago
      How so? I take free banking like historically practiced in Canada, Scotland or Australia as the model here---they dealt very will with fractional reserves. I don't see how introduction of other currencies would be a problem? Please explain.
apaprocki 10y ago
I think there have been similar things in practice for a while. The real difference here is just the open blockchain to simplify accounting. Even though there are probably lots of little details that are different, this seems really similar to things like XDRs: https://en.m.wikipedia.org/wiki/Special_drawing_rights https://en.m.wikipedia.org/wiki/Special_drawing_rights
jokoon 10y ago
What's the point if it's not public? Wouldn't it be faster to just wire money as it's being done currently?
- sp332 10y ago
  Blockchains are good for preventing double-spending without a central authority.
- smileysteve 10y ago
  Wiring money is exactly what they are trying to do. The point is that banks use independent ledgers that have no guarantee of consistency and in some cases dependent of manual synchronization.
dang 10y ago
Ok, we've incorporated your point and another user's to try to make the title above less misleading. Thanks!

pzone 10y ago

Yes, this is extremely misleading. If a few banks want to replace their current transaction settlements with a blockchain protocol tailored to their needs, that's great, I hope it works out for them. Worrying that this blockchain isn't "open" or "democratic" like Bitcoin is just silly. It's like worrying about the fact the NYSE hasn't open sourced its internal code.

dogma1138 10y ago
Bitcoin isn't exactly "democratic" also, the size of the blockchain the the difficulty of the mining pretty much means that anyone who wants to get into the business would have to invest huge sums of capital to get into the game. The rest of us would have to use intermediate parties that would handle everything for us.
- pzone 10y ago
  Right. I use my credit cards and simply trust Chase/Citi/etc to provide the services they've promised. I use a website to do Bitcoin transactions and simply trust they are not going to run away with my money. So there's even less of a justification to complain that this system is controlled only by powerful players.
- repomies691 10y ago
  Hmm, currently the block size seems still manageable to me (80GB). Also you can turn on pruning so you need to store only the amount you want, and you can still use all the major use cases. You don't need to mine bitcoin to use it. It is still very cheap to use bitcoin still as an independent node. You can use those gateways to make your life a bit easier, but from cost perspective I think they are not needed.
  - dogma1138 10y ago
    These things are eventually going to exponentially increase, if Bitcoin gets to even 1% of the global payment cards daily transaction rate you aren't going to be running nodes on your own. Now you can still get into running a node and even doing mining however as time passes the barrier to entry increases exponentially and considerably faster than the capacity of individual hardware so you can't even really catch up.
    - wongarsu 10y ago
      >These things are eventually going to exponentially increase While bandwidth and storage costs exponentially decrease. Handling the entirety of VISA's volume with bitcoin transactions would require at least a 4Mbit/s connection (about double for peak times). That's not exactly out of reach of normal consumers. In that scenario the blockchain would grow bigger by 40Gb each day, or 15Tb per year. If you want to store that all, with current hard drive prices that would be $40 per month for new hard drives. Of course if you're not interested in looking up other people's random transactions, most of the data can be pruned as soon as the coins are spent elsewhere, giving you barely any blockchain size increase. And all of those resources get cheaper all the time. Resource requirements aren't a fundamental problem for bitcoin. The block size that limits it to ~0.3% of VISA's volume, and the mining system that favours a near-monopoly for the person with the lowest operating costs are far bigger problems.
    - louprado 10y ago
      "If Bitcoin gets to even 1% of the global payment cards daily" Visa alone processes between 2000 and 10000 transactions a second. Bitcoin is architected to handle 7 transactions per second. From my understanding this limitation can not be overcome by adding more miners, it is by design.
  - uncletammy 10y ago
    I think you mean "blockchain" rather than "block size".
- galdosdi 10y ago
  As an aside, I think people use intermediate parties unnecessarily way too much with bitcoin. Yes, mining it yourself is not feasible so you do need one to get ahold of bitcoin and to get fiat currency back from bitcoin, but you don't need (and maybe shouldn't use) one just for storing bitcoin, paying with bitcoin, receiving bitcoin. That's the whole damned point of bitcoin, but then people go and store it with mtgox anyway even though they're not about to exchange it, and are bummed out when mtgox loses their money. It sucks, but it's funny that half the point of bitcoin is making this situation impossible, if you bother to do it right (Of course, if you trust a third party more than you trust your own ability to not lose/forget your private key, then I guess it makes sense, but you're trading a lot off in exchange for that)
- 20yrs_no_equity 10y ago
  You don't need to use an intermediate party to use bitcoin. You can submit transactions directly to the network from your wallet software without running a miner. The miners secure the network, but the nature of the system is that they are trustless, so long as %51 of them are honest. By having the cost of mining being high, that makes the cost of an attack on the network high. Which aligns the miners incentives (Eg: keep the price of bitcoins from going to zero as it would if the network were compromised) aligned with yours. The blockchain is a bit large at 80GB but you don't need to download the whole blockchain to run a wallet and make transactions there are "light wallets" that will do this, and currently in testing is an even more efficient method of doing this called segregated witness.

dogma1138 10y ago

This doesn't exactly act like a proxy currency (there would be legal issues with that). Banks are really interested in using blockchain like technologies for settlement and clearinghouse related tasks, it would allow them to have near real time settlements without much effort and could both reduce their liability and costs. Both concurrent and correspondent banking has pretty big liabilities banks have to keep standing accounts with funds to facilitate, these funds are pretty expensive for the banks to keep since they both do not earn any interest and they are not fractional. Banks are also tied to fairly expensive transfer schemes (ACH/FSP) and service providers like SWIFT because a lot of the fast transfer schemes have limited membership (there are only 10 clearing members in the FSP for example) most banks end up paying pretty large transfer fees for agency banks that perform the transactions for them. Having the ability to set up blockchains that could easily be extended which would provide both the messaging/transfer functionality and the ledger that can be used for account settlement and consolidation between banks, as well as being extensible enough to allow multiple banks to join in either to the main blockchain or to sub-chain makes all the sense for them. Banks however would not and for the most could not use "digital currencies" for their day to day dealings, the only currency they are allowed to "create" is their respective national currency.

eru 10y ago
> [...] the only currency they are allowed to "create" is their respective national currency. Which jurisdictions are you talking about here?
- dogma1138 10y ago
  If bank start creating their own currencies it is very likely to conflict with the fractional lending model used by virtually all modern banking systems.
  - eru 10y ago
    How so? I take free banking like historically practiced in Canada, Scotland or Australia as the model here---they dealt very will with fractional reserves. I don't see how introduction of other currencies would be a problem? Please explain.

apaprocki 10y ago

I think there have been similar things in practice for a while. The real difference here is just the open blockchain to simplify accounting. Even though there are probably lots of little details that are different, this seems really similar to things like XDRs: https://en.m.wikipedia.org/wiki/Special_drawing_rights https://en.m.wikipedia.org/wiki/Special_drawing_rights

jokoon 10y ago

What's the point if it's not public? Wouldn't it be faster to just wire money as it's being done currently?

sp332 10y ago
Blockchains are good for preventing double-spending without a central authority.
smileysteve 10y ago
Wiring money is exactly what they are trying to do. The point is that banks use independent ledgers that have no guarantee of consistency and in some cases dependent of manual synchronization.

dang 10y ago

Ok, we've incorporated your point and another user's to try to make the title above less misleading. Thanks!