4 ms·
Another blockchain solutionist in search of a problem to solve. Finland, 15 years ago: free bank transfers from individual to individual, money appears on the
by wott 9y ago
Another blockchain solutionist in search of a problem to solve.
Finland, 15 years ago: free bank transfers from individual to individual, money appears on the bank account of the recipient in a matter of minutes or even just seconds, including the time it takes for him to check on the cash dispenser interface or the bank website. Regular average Joes' banks. No blockchain in sight.
- hanklazard 9y agoI've been thinking that bank transaction reconciliation is one of the best use-cases for DLT. No need for a crypto coin, each financial institution could just hold a ledger copy. R3's Corda is being built to plug in directly to legacy SQL systems, and its founders have lots of experience in finance. If anyone is interested, I found this epicenter podcast interview really interesting: https://youtu.be/3YTSwB5UrEI https://youtu.be/3YTSwB5UrEI
- mrb 9y agoSigh... I never said the problem can be solved only by a blockchain. But the problem does exist in some countries and the blockchain does solve it. Also the problems of the Finnish system that cryptocurrencies solve are obvious: your system has low-ish daily/monthly limits, and your system only works with 0.05% of the world's population (namely those with a Finnish bank account). In an era of European and international trade, that is ridiculous.
- dsacco 9y agoBut why use a blockchain instead of a SQL database? What concrete benefits does it provide and why do we want those benefits? It's just a data structure, and it's a great one at that, but I really don't see how it fits here.
- mrb 9y agoA blockchain is necessary for decentralization, which brings 3 main benefits: 1 - Censorship resistance. There is what I call big and small censorship. It can be an authoritarian government seizing the bank account of a journal supporting the opposition, or it can take the form of Paypal freezing someone's account for the wrong reasons (a few examples: https://news.ycombinator.com/item?id=1803488 https://news.ycombinator.com/item?id=1803488 or https://www.dailydot.com/layer8/paypal-protonmail-freeze/ https://www.dailydot.com/layer8/paypal-protonmail-freeze/ or https://news.ycombinator.com/item?id=11048131 https://news.ycombinator.com/item?id=11048131). Cryptocurrencies solve this by allowing users to be in absolute control of their cryptocurrency addresses and keys. 2 - Irreversibility of payments. This helps merchants avoid fraudulent credit card chargebacks: once funds are received they are spendable right away. No one can reverse payments. No system other than blockchains offer irreversibility. 3 - Permissionless. You don't need anyone's permission to participate in Bitcoin or other cryptocurrencies. Banks sometimes refuse to let people (especially the poor) open bank accounts. Because they don't have enough for an initial deposit, or have a bad credit history, etc. By contrast, anyone can be a cryptocurrency user as long as they have a $20 Android smartphone.
- jimktrains2 9y ago2 - Irreversibility of payments. This helps merchants avoid fraudulent credit card chargebacks: once funds are received they are spendable right away. No one can reverse payments. No system other than blockchains offer irreversibility. Flip side is that customers can't reverse payments when they're victims of fraud.
- mrb 9y ago"customers can't reverse payments when they're victims of fraud" No different from cash, yet we tolerate this aspect.
- jimktrains2 9y agoIf we're comparing against cash, then it's difficult to fraud the vendor as well, so the originally mentioned case is moot as well.
- mrb 9y agoNot moot. My point is pretty simple: payers tolerate irreversibility, recipients like irreversibility, hence Bitcoin can (and does) work.
- jimktrains2 9y agoYou're missing my point. I'm not saying btc won't or doesn't work, but that there is a flip side to the no reversible transactions, and one that merchants already deal with.
- tehlike 9y agoEscrow would kick in here.
- dsacco 9y ago> A blockchain is necessary for decentralization. Every blockchain initiative being explored by financial institutions is not decentralized. Unless you're tossing out the banking system to directly transact with other parties using digital currencies, in which case: good luck, I don't have a rebuttal except to say I doubt that will become mainstream enough to be ubiquitous and convenient, but maybe I'm wrong. > 1. Censorship resistance. I don't personally care about this; do you think most people do? > 2. Irreversibility of payments. That's admittedly cool, but I personally don't want that either. Great for vendors I suppose, but I like being able to e.g. issue chargebacks with AmEx when there is legitimate fraud. I don't think there is a technical solution that satisfies both buyers and sellers and mitigates fraud on both sides. > 3. Permissionless. I don't really understand this, I have accounts with several banks that didn't run credit checks before I opened an account. They did an identity verification, but they didn't care about my creditworthiness for a checking account. Moreover, these were national banking chains with online account signup and approval, so it isn't limited to regional availability. Other than that - I agree permissionless is a cool feature, but I don't particularly want to transact with people in a permissionless system. The natural extrapolation of a permissionless system is one in which there are fewer safeguards for e.g. verifying a user's identity, and (speaking as someone who works in cryptography research here), I don't think schemes that enable two parties to transact without mutually trusting each other are a worthwhile problem for a technical solution in the real world.
- kwhitefoot 9y agoSimilar here in Norway although the transfers are batched when the recipient is at a different bank. Batches run three times a day. Transfers to accounts at the same bank are instantaneous.