For some Nostr applications, it is important to have data that can be verified, but that you don't want to be published (eg SIGit agreements, medical records, or internal application data). One way to achieve this is by sealing (kind 13) but this has some drawbacks - such as the fact that only the sender and receiver can decrypt / verify the content.
NIP-59 suggests that removing the signature is one way to prevent events from being published. But this also stops it from being verified.
With this in mind, we would like to #asknostr for feedback on the following idea.
Using negative kind numbers for private data
Negative kind numbers MUST NOT be published to relays, and relays MUST NOT accept negative kind numbers
Thoughts?
Cool idea! It makes sense to reserve all negative numbers for private kinds since there could be any number of private kinds.
However, I wonder if there are any other things that also need their own space. Maybe a binary prefix (+ or -) limits the number of orthogonal kinds a bit much.
In Bitcoin we use `bc1` to indicate that an address is a bech32 main net address and `tb1` to indicate that it's a bech32 testnet addresses.
I wonder if we can introduce similar prefixes for nostr notes or signatures...
You've been working on medical data @Vitor Pamplona, right? Any thoughts on negative kinds?
Ive been personally using negative numbers to sign things, i think its a decently elegant solution
As of now, kind is an unsigned int. So negative numbers cannot exist until we change how nostr encodes event kinds, which might change everything.
But GiftWrapped & Sealed medical data is a thing. We have been doing a bunch of experiments. It's hard to crack the growth of it, but definitally possible.
Well that answers that, thanks!
#asknostr
nostr:nevent1qqsqqmq6p8feg8850xc2xwkf48wg4m6t4h7ul5qq0hum9xptvasf09gpr9mhxue69uhhyetvv9ujumn0wdmksetjv5hxxmmd9upzqf6g0jtqpvttyjsmlvz3nnly5hg6mpy4nc7vuhtd02vafpnq58mcqvzqqqqqqyt4cfaj
Oddbean new post about login | logout