yeah i'd be happy to scale based on payment but also there's this weird gray area between actually running a viable business and hobbying this out on tips or micropayments or whatever. i like some kind of peering approach because then i could participate without having to do either of these
Yeah, agreed generally what you're saying. I'm doing this for fun. I wouldn't be able to develop the relay code I'm working on without actually operating a relay with some kind of load, so I'll put a cap on connections or data some point like others are doing.
No at that point your phone itself will be Damus
I’ve seen profiles flip-flop like that across other clients.
Updating relays is kinda tricky for clients to get right.
When you remove relays, a good client should first write to the profile with the full set of removals to all removed relays.
When a client reads profiles, it should make sure it caches the created-at timestamp and always let newest win.
Any latencies to get latest profile will also make your experience wonky.
Do you see a lot of limit-less firehose queries? I’ve been thinking about auto-limiting users/clients that request the same big query repeatedly.
I’ve been thinking relays should be able to redirect clients to download older (and compressed) event data from “firehose” subscriptions (i.e. [.., {}]) from a different url/endpoint — which could then be cdn cached etc.
😎 how many active users? do you fulfill limit-less subscriptions fully (i.e. all of history?)
Yeah -untether’s /q endpoint is really just for human admin/operator observability but should it be an restish api :thinking:
I might move to support https://github.com/nostr-protocol/nips/blob/2fa78a8097745139be5299c9f6117049f7d4a074/22.md but only restrict far-future events. At the end of the day, though, I don't see how clients can't avoid having to deal with future events....
Am I the only relay that still supports future-dated posts?
Am I the only relay that still supports future-dated posts?
what latencies are you seeing?
my tiny relay is holding up decently but I’ll prob need to upgrade the machine soon
https://github.com/ACINQ/secp256k1-kmp -- fme way faster than pure JVM impls
why not sell nostr to elon musk
I’ve been on vacation. What have I missed in nostrland
Thinking...
Thinking...