4 ms·
> If apple.eth is held by a squatter, should courts be able to override the protocol to send it to trademark holder Apple? In the end this can get resolved by r
by acdha 4y ago
> If apple.eth is held by a squatter, should courts be able to override the protocol to send it to trademark holder Apple? In the end this can get resolved by regular dispute mechanisms but without compromising decentralization and integrity of the protocol.
How exactly do you think this works? With DNS, there's a trusted third party who can help recover from errors — if you lose access, you might have to prove your identity but it's not a hard process. If someone squats your domain or steals it when they compromise your computer, the registrar can sort it out or will respond to a court order.
With ENS, the answer is you're screwed if you lose the key or whoever has it currently isn't responsive. Maybe you can track them down and get a legal action strong enough that they honor it but that's a lot more work and has a lot more failure modes which look like the apple.eth service having to move to apple computers.eth because the system is designed to fail hard. Very few people are going to want to build on something like that because it means that even a trillion-dollar corporation is one phish away from a massive disaster.
- web3isgoing 4y agoI understand how DNS works. The design goals of ENS are not the same as DNS. There is no "trusted third party" in ENS because the system aims to be trustless, decentralized, ownerless, and censorship-resistant. Even in extreme scenarios like a US sanction, tornadocash.eth domain is still functioning[1] but tornadocash.com has been revoked by Google registrar[2]. Not everybody in the world feels that the current US government administration should be able to dictate and censor worldwide internet protocols in this way. > Maybe you can track them down and get a legal action strong enough that they honor it That's fine, it's how our legal system works. Using a bank as an example: they can eject you as a customer if you do something they think is criminal, even before you have had a fair trial that might prove you innocent. ENS and crypto both ask: what if you create protocols where users actually do have complete ownership over their assets, and no single entity has more control over the ledger than another? [1] https://etherscan.io/enslookup-search?search=tornadocash.eth https://etherscan.io/enslookup-search?search=tornadocash.eth [2] https://www.whois.com/whois/tornado.cash https://www.whois.com/whois/tornado.cash
- acdha 4y agoI’m aware that it’s a thought experiment but it’s weird to have that combined with unhappiness that few people are using a system which is designed to prevent real world problems from being solved. I’d also lay off the marketing-speak and think about what tangible benefits ENS provides in your example: yes, tornadocash.eth resolves but most people cannot use the service safely and even your link shows the content as blocked – not exactly a meaningful anticensorship advance. If you were actually concerned about censorship, an expensive service which leaves a public audit trail for prosecutors would not be the optimal starting point.
- web3isgoing 4y agoThe real world problems are still being solved, just off-chain. I made a similar example with Mastodon: no central entity controls Mastodon, so x@x.com and x@x.xyz might be owned by different people, one impersonating another, and you can't just "complain to Mastodon" to fix it. Tornado cash: the domain is still resolvable to an IPFS hosted website[1]. As far as ENS is concerned as a name service and URI resolving mechanism, it is working great. This doesn't mean the smart contract that facilitates tornado cash deposits is working that well: if the US says you will be criminalized for touching a smart contract, then nobody will want to touch it, and tornado cash depends on many people using it to provide anonymity. > If you were actually concerned about censorship, an expensive service which leaves a public audit trail for prosecutors would not be the optimal starting point. If you are referring to tornado cash, I think you are mistaken. [1] https://twitter.com/liamzebedee/status/1578127982173908992 https://twitter.com/liamzebedee/status/1578127982173908992
- acdha 4y ago> The real world problems are still being solved, just off-chain. This is incorrect and the example shows why: > I made a similar example with Mastodon: no central entity controls Mastodon, so x@x.com and x@x.xyz might be owned by different people, one impersonating another, and you can't just "complain to Mastodon" to fix it. In this case, it might not be a problem — many people share the same names and a decentralized system supports that in a way which a centralized system like ENS does not. If it is a problem, you can complain to the operator of that particular instance and if it's an actual crime their domain registrar or hosting company, etc. If you lose your credentials or someone steals them, you can use the same mechanism to correct it. In the case of ENS, you not only do not have any of those mechanisms but the system is designed to prevent them. If you lose control of your keys, you lose your name for ever. If someone squats on your name, there's no third-party to turn to short of filing a lawsuit and hoping that the person complies with the eventual outcome. Re: Tornado cash, that's kind of the point – the smart contract is risky for people in multiple large economies to touch, and IPFS is no more resistant to take-down notices than any other distributed hosting option since it's built on top of the normal internet and governments have plenty of experience contacting network owners. Having the name up gives you something more reputable looking than using a DNS TLD under a different legal authority but it's not meaningfully making the system more available.