4 ms·
The equivalent to the relay in that case is just the browser. If control of the browser passes to someone else - say because the dev is in jail and the server o
by troad 4y ago
The equivalent to the relay in that case is just the browser. If control of the browser passes to someone else - say because the dev is in jail and the server or signing key is seized - you now have a silently-self-updating piece of software running code from an unknown third party. This seems like a far more catastrophic problem than losing control of a relay domain.
- ETH_start 4y agoThe browser is client side code, that runs independently on millions of people's computers. It is not a centralized relay, and control over someone's browser can't be "passed to someone else". On Chrome, all updates have to be manually approved by the user - it is not "silently-self-updating". A developer going to prison, and a relay failing because he couldn't update the registration, is a far cry from a developer's signing keys being passed onto a malicious replacement who then includes removal of Web3 functionality in a browser update that users approve. >>This seems like a far more catastrophic problem than losing control of a relay domain. That's completely irrelevant, as the risk you describe is just as present in Web2. Risk is strictly reduced by removing the reliance on Web2 relays, the risk of browser hijacking notwithstanding. >>Web3 boils down to "we've decentralised it, which is inconvenient, so you can find links to all of the decentralised bits in this one convenient central location, so just go there". I mean that's what the early Web was like, with people going to centralized mainframe computers to access it. Any technology will take time to deploy, and its full benefits will be largely unrealized while it has low adoption rates and primitive tooling.