Comment by web3isgoing | Hacker News Reader

Comment by web3isgoing | Hacker News Reader

Parent Full thread

ENS avoids this dispute mechanism by design, to avoid the need for a central third party.

> If the holder is using it maliciously to scam people by tricking them into thinking it's the real Apple, then yes!

Try applying your logic to Twitter and its current situation. Is it a good thing that accounts impersonating others are being suspended from the network?

Mastodon does not have this same centralization, so @teraflop@teraflop.xyz could impersonate your @teraflop@teraflop.com account. There is no centralized entity that controls the whole Mastodon and ActivityPub network, which I think is a good design decision for a less centralized protocol.

And see my sibling comment:

https://news.ycombinator.com/item?id=33626408

Local namespaces are fine, even desirable since very few names are globally unique.

The problem with ENS is that it apes the federated DNS design without the oversight mechanisms. In the example you gave, an impersonator is subject to the rules of their Mastodon instance and the operators can handle things like lost credentials or an account compromise, and if a particular host is not a good custodian many people will block the entirely. ENS has none of those corrective mechanisms and I suspect that they’ll slowly be added over time to get usage.

I could be wrong, but I do not believe ActivityPub spec actively outlines suspension rules, instead letting servers decide how to moderate their content. Same with clients on top of ENS and the blockchain.

These moderations already exist on services built on top of ENS, like tornadocash.eth.limo which fails to resolve because the .limo domain is following US sanctions.

> In the example you gave, an impersonator is subject to the rules of their Mastodon instance

Just to point out, the impersonator can self-host, so Mastodon protocol has no course to expel them except for each other instance to suspend the impersonator one at a time, or to come to consensus on large blocklists.

Reply on news.ycombinator.com