https://en.wikipedia.org/wiki/Society_for_Worldwide_Interban...
See also:
https://en.wikipedia.org/wiki/Clearing_House_Interbank_Payme...
I worked on a payment system using the UK's Faster Payment Scheme a while ago. The SLA for latency is two seconds end-to-end, and is usually much faster than that, was simple enough for a gang of underpaid COBOL programmers to implement it successfully, and didn't have a mechanical error rate worth bothering about. In January it moved 186.1 million payments and 161 billion pounds:
http://www.fasterpayments.org.uk/statistics
FPS is UK-only, but there's nothing about the technology or the commercial and legal machinery that wouldn't work internationally. If SWIFT is really as bad as you say, it could be replaced with a better instance of the same technology; it doesn't require a haunted git repository to fix it.
lol, thanks for that.
All the big players have got real-time feeds they're sharing with each other to keep track of who's made what transactions, and they're constantly reconciling them against each other, to make sure that everyone's looking at the same picture. There are also daily reports summarizing the previous day's activity, and they are also reconciled as a second check. The proactive parties work together to sort out any discrepancies they find as soon as they find them.
None of this is strictly necessary, because there is a central source of truth that you can rely on. But it tends to operate much too slowly for a lot of people's needs, and mistakes do happen, and they can be costly, so it pays to be proactive. Meaning that, in general, nobody's actually relying on the central authority to keep this organized.
The actual real-time feeds are log-structured: You can add new records, but not delete old ones. If you want to reverse a transaction, you just add a new entry to the ledger to back it out.
In that sense, it works a lot like how a blockchain would work. The only difference is, nobody's bothering with all that crypto stuff, because all it would do is make the whole thing more expensive. Like, literally, that's the only impact I can see. It won't help with preventing errors - when mistakes happen, it's invariably because of an error at the boundary of the system, and fiddling around with hashes of the previous message when sending new ones does nothing to guard against that. All it would do is slow both the throughput and latency of transactions by introducing more spots where synchronization must occur together with a bunch of ancillary computational load.
In a world where time is money, that sure sounds to me like a $100 solution to a 1¢ problem.
https://en.wikipedia.org/wiki/Real-time_gross_settlement
For euros, there's TARGET2:
https://en.wikipedia.org/wiki/TARGET2
Daily average of 342008 payments, value 1.7 trillion euros:
https://www.ecb.europa.eu/paym/target/target2/facts/html/ind...
Blockchain does exactly this without everybody having to sort out discrepancies. Everybody getting the same picture is built in. Just the possibility of there being discrepancies in the ledgers you described would have me looking for a better way.
I would much rather have the option of connecting to a decentralized feed that is the same for everybody regardless of resources.
It's all pretty highly regulated, and you'd need a license before you're allowed to even think about hooking directly into an exchange. If you've got lowly peasant money to work with, it would be much more cost effective to go through the retail channels, just like every other person who isn't a corporation.
> Blockchain does exactly this without everybody having to sort out discrepancies.
No, it doesn't. The nature of the discrepancies that cause trouble in practice, and the ways they sneak in, are simply not things you can solve by chaining hashes and throwing a proof-of-work algorithm into the mix. The very suggestion puts you one blue suit and some pointy hair away from being able to star in a Dilbert strip about trying to use buzzwords to solve problems you don't understand.
Financial markets are heavily regulated for a good reason, Bitcoin need to find a good way to "clone" that if it want to be used for actual finance.
Errors can come from outside the system, any kind of consensus protocol cannot solve that. So cryptocoins do not have any particular advantage here.
That is exactly what blockchain is. It is literally doing that.
> None of this is strictly necessary, because there is a central source of truth that you can rely on.
But who is running that? And why do you trust them not to change things around every so often?
> it works a lot like how a blockchain would work
Except there's no central source of truth because no one trusts anyone. Which is why its so strong. There's no need to trust.
> The only difference is, nobody's bothering with all that crypto stuff, because all it would do is make the whole thing more expensive
That's what makes blockchain immutable. You can't alter it without an enormous amount of resources. And even then, the actions of the rest of the network work to discourage that sort of thing. In PoW, everyone else is checking the non-altered data, and in consensus, everyone else has already agreed on non-altered data.
It isn't: it's a chain of hashes with signature. Those predate blockchains that do things like wasteful mining. An example was Surety's timestamping service. A hashchain using standard primitives is way less wasteful, supports higher transaction volume, is cheaper, and can take advantage of hardware acceleration in client and server devices.
One of reasons I oppose blockchain tech is that you get better cost-benefit ratio out of high-performance, centralized protocols with decentralized checking. We also have some of that mathematically verified for correctness down to assembly. So, they will be more secure than complex, decentralized protocols.
Ain't no judge gonna throw up is hands and go "well, the smart contract says so, my hands are tied!". Blockchain doesn't solve anything where there is a functioning legal framework.
You can't use technology to solve a problem that isn't a technology problem.
You still have to trust the other party will abide by the results of the smart contract.
If there is no legal framework they could just shoot you and have done with it, if they didn’t like the result.
Smart contracts and blockchains are still laws, just laws for people who don’t like some of the existing ones. Numerical rather than societal.
If there’s no rule of law, then blockchain laws can’t exist either.
Central bank?
> Why should the other bank trust that database?
Because if that database fails, it is likely that the failure is one of the smaller problems of the bank?
Goodwill is an asset that companies care about.