In writing, NIP-95 sounds good. It allows our images to be decentralized! We don't have to use centralized services to host images and video! This is what we've wanted from the beginning, when realizing that our text was decentralized, but our images and videos were not.
In practice though, NIP-95 overloads relays with large amounts of data. Relay operators signed up for hosting just text messages. Hosting data files as text inside json was not what they intended to do. This will drastically increase their storage footprint and server costs.
I don't think NIP-95 is bad as it solves a real problem that we do have, however rolling it out needs a better implementation path instead of via brute force. This is the type of change that has consequences and all must be considered and discussed by client developers, relay developers, and relay operators - both paid and free.
Thread
Login to reply
Replies (6)
Want to organize a nest about it? I am happy to join and answer any questions
And Iβll further centralize the relays, only the ones with the money to hosts image and text would be standing, even if there is an opt in to turn images off, most people would prefer image hosting relays. Nada bueno.
Would be cool if there was a way to share the image between relays without hosting the full file, kinda like torrenting but with letβs say pixels of an image. That would reduce the amount of data is stored while allowing the relays to show the image in full.
Iβm not a dev whatsoever π and my line of thought might be wrong as well
Pay sats to relays to upload images using nip-95 ?
View quoted note β
If you want total centralization of Nostr network, then NIP-95 is perfect.
If node also don't use NIP13, the centralization is done in an afternoon by a bad actor or an asshole.
Wow.. it's a good news and real great introduction for protocol.. π§‘π©β‘π
#[0]
Very soon everything will be decentralized, and we will have the freedom we've always desired for.