Not for long. nostr:npub16c0nh3dnadzqpm76uctf5hqhe2lny344zsmpm6feee9p5rdxaa9q586nvr has confirmed they will add a fallback where the client reads directly from relays in the case of a caching server outage.
A caching server is not a bad thing, especially as Nostr grows. Primal will always be the fastest client with the largest user base. The other clients are ugly and slow.
A caching server removes the entire point of having relays. That is why everyone is working on outbox.
everyone except the homos at primal
It really doesn’t. The relays are still decentralized. The caching server greatly reduces load on the relays making it possible for more people to run relays. The caching server improves responsiveness and many other things.
You can have both. You could have a caching server and have the client read directly any relays that aren’t cached
But you are reading from the caching server which can control what you see and in that case you do not need a relay
Primal has censored content multiple times
Also, running relays at production scale (with nostr.land) costs barely anything even though it is the default in a lot of clients
Are you retarded? What does the caching server read from? You still need relays.
Show me who they censored, and why. I’ve asked several people this and none of them have given me proof of anything being censored.
You cry about centralization and then say the solution to scaling problems is centralizing relays. The whole point of relays is that you can run your own and post content to it so no one can censor you.
A client that caches popular content and allows direct reading from unpopular relays would be more performant, and it would reduce energy and data usage.
You don’t even have to use the primal caching server. It’s open source.
