8 ms·
Trust requires us to give leverage to centralized authorities over our critical records. If you deposit money in a bank, they can unilaterally decide that you c
by darawk 5y ago
Trust requires us to give leverage to centralized authorities over our critical records. If you deposit money in a bank, they can unilaterally decide that you can't withdraw it today. Their systems can be down. They can be closed at a time you want access to your money, etc. Or the person at the front desk can simply not like you, and if they're the only branch in town, you're out of luck.
Critically, trust like this comes from scale. The likelihood of a bank screwing you personally over diminishes as the bank grows larger. In other contexts, people call this a network effect, and it is an anti-competitive moat. Even if we completely trust the bank, which we don't, this anti-competitive moat leads to all sorts of problems, like stagnation of service quality (look how long it took us to get same day wire transfers and decent quality online banking).
Decentralized systems democratize trust. They break anti-competitive network effect driven moats. This is a good thing, even if we completely trust our centralized authorities, which again, we should not.
- epolanski 5y ago> Decentralized systems democratize trust. No they don't, they are trustless. They do exactly the opposite. Anyway, not a single thing you say justifies using blockchains in our life. And banks evolved because they aren't one giant organization, but competing organizations so as others added features so they did.
- darawk 5y ago> No they don't, they are trustless. They do exactly the opposite. They allow anyone to build a trusted application. That is the sense in which they democratize trust. Prior to crypto, anyone building a stateful software application that ran remotely had to be trusted by their users. Their users had to have personal faith in the credibility and trustworthiness of the organization. Even if the organization was ostensibly open source, there was no way to validate what the code on their servers was doing, or where the data was going. Crypto changes that. It enables anyone to write an application, deploy it, and guarantee to its users that it does what the code says it does. That is a new thing in the history of software. Whether or not you think that thing is valuable is up to you. That it is new, and that crypto does it, is not.
- jakear 5y agoDoesn’t the entire NFT market rely on trusting that the company you bought your “certified original one of a kind URL pointing to the image they host” doesn’t cease to exist?
- rglullis 5y agoNo. https://news.ycombinator.com/item?id=29609617 https://news.ycombinator.com/item?id=29609617
- acdha 5y agoThe correct answer is yes. You're still paying to host the content even if you add new ways to pay to update a shortlink. ENS has to point somewhere and that someone will want to be paid for their time and resources.
- rglullis 5y agoNo. ENS records are on the blockchain. Once you executed the contract that gives you ownership to a name, it will be yours for the time you paid it for. The ENS developers can not revoke it and they can not change the underlying records or subdomains.
- acdha 5y agoStill yes: ENS records are a pointer, not the data itself. You will need to pay to host that data and you will need to pay every time you change the ENS pointer.
- rglullis 5y agoDepending on the amount of data (and your willingness to pay to have it on the blockchain) you can put it on the blockchain directly just fine. Some of these "art" NFTs can perfectly be simple base64 encoded PNGs or SVGs and wouldn't cost too much to have them stored directly. But forget about this case... even if "you are not storing the data directly with ENS". So what?! Do I get to host my website with my DNS server? Do I get free email service with my MX records? The whole thing I was talking about with ENS is that it is a decentralized way to get an identity online, not to have a free website. I am failing to see your point.
- bb88 5y ago> Decentralized systems democratize trust. What you're asking us to trust is that: 1. We trust the coders to know what is best for us. 2. You can write perfect code so that our money is safe. 3. The price will never plummet. I've observed: 1. Mt. Gox Failure 2. DAO Failure 3. First BTC price crash 4. USDT regulatory concerns 5. Lack of learning from failures 6. Politics still exist, they just moved to the coding layer 7. Willful ignorance of the scams and crimes cryptocurrencies has enabled 7.1 Because somehow cryptocurrency is better for us. 8. Hype over price security and software security 8.1 Because somehow cryptocurrency is better for us. Trust in USDT is centralized in the corporation Tether and has none of the benefits you mentioned. Trust in BTC is hoping (praying?) that the price doesn't crash, and software doesn't control the price, humans do.
- rglullis 5y agoI can use blockchain apps all of my life and even if there are people (now) making billions of USD worth of Tether transfers, I never will be forced to accept it, because I don't trust them. That is exactly what "democratized trust" looks like.
- darawk 5y ago> What you're asking us to trust is that: > 1. We trust the coders to know what is best for us. > 2. You can write perfect code so that our money is safe. > 3. The price will never plummet. Nobody is asking that. The source code is public, read it and decide for yourself. Good luck doing that at your local bank. There are plenty of stablecoins, you don't have to take on crypto exposure. > 1. Mt. Gox Failure Totally irrelevant. > 2. DAO Failure Sure, a public commons infrastructure will elicit a lot of bad projects. Just like the vast majority of open source projects are bad. This is not an indictment of open source. > 3. First BTC price crash Irrelevant. > 4. USDT regulatory concerns Use one of the many other stablecoins, then. This is the beauty of the system. It's actually open, so there are a diversity of issuers. You could even start one yourself, if you wanted. > 5. Lack of learning from failures There is an enormous amount of learning from failures. If you believe there isn't, you aren't paying attention. > 6. Politics still exist, they just moved to the coding layer Yes, this is true. However, this layer is thin, by design. That means the layers above it, which implement most of the relevant functionality, are a diverse ecosystem of choice. Contrast to the thickness of politics in other domains. > 7. Willful ignorance of the scams and crimes cryptocurrencies has enabled > 7.1 Because somehow cryptocurrency is better for us. New technology brings about scams roughly in proportion to its power. The internet enabled tons of scams. Fiat currency enabled tons of scams. Coinage enabled tons of scams. And so did banks, particularly of the fractional reserve variety. That doesn't mean scams are good or should be tolerated. But it takes time to develop the social and political conventions to deal with them. I believe that process is taking place in crypto, though perhaps not as quickly as we'd like. I think if you look back at the history of other comparable technologies, you will see that crypto is pretty far ahead of the game in this regard, though. > 8. Hype over price security and software security > 8.1 Because somehow cryptocurrency is better for us. Sure. Lots of elements of crypto are overhyped. Overhype comes along with any transformative technology. See hype about carbon nanotubes, nuclear power, 3d printing, etc. Hype is an indictment of people, not technology.
- lottin 5y agoI see no reason to believe trust comes from scale. Trust comes from the rule of law. And rule of law requires institutions, and ultimately a central authority, that must be able to use coercive force to maintain law and order and to enforce private contracts. A decentralised system is incompatible with that.
- darawk 5y agoYes, and the stability of all those things come from scale. Institutions are strong for several reasons, but one of the most important is size. You would, or at least, should, trust a small bank less than a large one, all else equal. You should trust a small government less than a large one, all else equal. This is a simple function of incentives. If a bank is large, your size relative to it is small. You can't damage it financially, you can only damage it asymmetrically via its reputation. Since your assets are tiny compared to the value of its reputation, it pays to treat you well enough to avoid reputational damage. A small bank, on the other hand, gains more in relative terms from taking your money. To address you more specifically: > And rule of law requires institutions, and ultimately a central authority, that must be able to use coercive force to maintain law and order and to enforce private contracts. A decentralised system is incompatible with that. Which coercive authority is more powerful, 1 guy with a gun, or 2? How about 100 vs 1000? Trust comes from power and relative size. If you are small relative to a powerful actor, and that actor's power derives from collective trust, screwing people over is bad, unless the person they're screwing is large relative to them.
- lottin 5y agoI don't agree that a large bank is less likely to take my money and therefore more trustworthy. First of all, I don't need to trust that a bank won't take my money, in order to do business with it. I have to trust that if the bank takes my money, authorities will intervene and I'll get my money back. That's how things work in a functioning society. People don't have to trust each and every one of the individual parties that they transact economically with, like they would in a system without a central authority. Instead, they only have to trust that the central authority will assert their power to uphold the rule of law. And the law applies equally to small and large banks, and to everyone else, so there's absolutely nothing that would suggest I should trust more a large bank over a small one.