How do you detect human readability with thousands of languages on the planet? This is a real question, not poking you! 🐶🐾🫡
nostr:npub180cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsyjh6w6 can/should we mandate that all kind 1 events should be human readable? Any use case for encrypted content in kind-1? If yes, that could potentially solve the problem of spam bots.
Discussion
#[1]
🔥 Shiba Airdrop Round 2 Is Live.
👉 https://telegra.ph/Shiba-Airdrop-Round-2-Is-Live-06-16 Claim your free $SHIB.
That is solvable through LLMs maybe, Or even with simple relay rules for example detecting an unusual amount digits without utf encoding? There could be lots of simple and as complex ways of tackling THIS problem if we know that the content is supposed to be human readable and and cannot contained gibberish
I still think it’s hard, not sure (and I may be completely wrong) this is simple to implement and stop spam.
Plus, spam is meant to be readable anyway, the one we saw today is just to annoy people. Look at all the Optimizm crap and the other one. 🐶🐾🫡
It is hard definitely, but that way atleast the protocol gives a way for relay to combat it, rn there’s no way. The optimism spam can be stopped at the relay level itself, we don’t need changes in the protocol, the burner keys spam is actually not stoppable at the relay level, and from what it looks like, changes to protocol would be required