Oddbean new post about | logout
▲ ▼
tyiu | 3 months ago (raw) | export | reply | flag +9 
 Looking for feedback on this PR to update the terminology on NIP-09 - “Event Deletions”. I’m proposing we call it “Event Retractions” instead when we actually mean “request for deletion”. Let’s reserve “delete” for when we actually mean physically deleting an event from a specific client or relay. There’s too much confusion stemming from overloading the term. Not all clients or relays need to support retractions, but the protocol describes how one could support it, so let’s use the correct words that explain it properly.

cc @rabble @Ava who have been vocal privacy advocates.

https://github.com/nostr-protocol/nips/pull/1425
▲ ▼
crypt0cranium | 3 months ago (raw) | root | parent | reply | flag 
 i like delete
▲ ▼
[ARCHIVED] Jay | 3 months ago (raw) | root | parent | reply | flag 
 Request to memory hole
▲ ▼
tyiu | 3 months ago (raw) | root | parent | reply | flag +1 
 OK, I changed the language a bit after some PR feedback. It’s now “deletion request”.
▲ ▼
Cameri | 3 months ago (raw) | root | parent | reply | flag +1 
 Relays publish an information document on whether they support NIP-09 (among other NIPs) but few clients show this information to users, so many can’t really make an informed decision on which relays to use or not.

NIP-09 also specifies that clients MAY choose to hide deleted events, which perhaps needs to be changed to SHOULD. Otherwise claiming NIP-09 support as a client means nothing for the user really. And even then, I am not aware of any Nostr client that let users know which NIPs are supported so there’s that too.

If anything, relays are likely doing a better job with NIP-09 than clients are. I have used a few clients and none allowed deleting events (they did honor other users deletions).