3 ms·
>My understanding is that many of the people who own crypto do so through a third party, so there is a layer of indirection. It's the difference between me havi
by cowtools 4y ago
>My understanding is that many of the people who own crypto do so through a third party, so there is a layer of indirection. It's the difference between me having cash in hand (money in my pocket I can directly use) and me having cash in the bank (I tell my bank to send money to someone else and they execute the transaction on my behalf).
This is quite true, and it is likely the largest problem facing cryptocurrency today is this custodial use of it (besides all of the get-rich-quick schemes). But at its worst like this, cryptocurrency is a non-proprietary inter-bank payment method that prevents double-spending between banks. It is still superior to something like zelle, paypal, or SWIFT so long as the fees are lower. If cryptocurrency was the primary means of inter-bank transfer, then it would be trivial for anyone to start a new bank that could inter-network with the rest of the banking system, so I expect banks would be a lot more competitive (including on matters of privacy and security).
>My mom has downloaded ransomeware before, so from that perspective, I think crypto has worse security properties. If transactions are executed indirectly, the security properties are theoretically the same as executing transactions through a bank and you are back in a system of trust. Furthermore if a "cryptobank" gets hacked, that money is not retrievable, while theoretically in a system of pure fiat, the money might not be retrievable, but
Hmm. yes this is sort of a complicated subject. But I'll just re-iterate a point here which I may not have made as clear earlier: that cryptocurrency allows you to establish different levels of trust/risk through the means by which you manage your keys. A lot of older cryptocurrency users who don't practice good opsec will use a hardware token to sign transactions. Another example of what you could do is use a multi-signature system that would make it so that multiple keys are needed to move your funds (for example, they would have to hack at least X of Y devices in order to move funds), or simply have multiple wallets and limit the amount that you have in each one.
And secondly, there are non-cryptocurrency ways of implementing different levels of trust/risk that you could integrate into the existing banking system, like chaumian cash or even just using cryptographic keypairs to authenticate transactions.
In other words, losses of cryptocurrency due to theft or fraud are not always all-or-nothing. The difference between cryptocurrency and the conventional banking system is that you can decide your level of trust/risk you want to take before you do a transaction, which includes the use of a "cryptobank" (which could be secure but have historically been very scammy compared to conventional banks, see Mt Gox, Celcius, etc.).
>the value could be refunded at the cost of devaluing the currency as a whole.
I am not sure that it's a desirable property that the rest of society can bail out banks like you're describing. I think in an ideal situation you would have some sort of free-market-ish sort of way to balance the risk vs reward of different security practices, whether that's users voting with their dollar or with a middleman like rating agencies or insurance. And those incentives basically require the bank and its customers to lose money when they get robbed (maybe through some middleman like insurance).
If you look at serious cryptocurrency exchanges like Kraken or Binance, there is a massive gap between "cryptobank gets hacked and loses some of their funds" and "cryptobank gets hacked and loses everything". They keep a lot of their funds on separate, air-gapped, offline systems, with the keys distributed between multiple people. Those aren't funds that you can steal with a normal cyber-attack: it would take pretty persistent social engineering akin to widespread corruption.