Replying to Avatar Andrew

Considering I open Damus and Instagram with similar frequencies, but Instagram feeds have way more images and videos, it would seem that it’s not media that uses up the bulk of bandwidth on nostr.

Rather, this seems to show that downloading tons of redundant data from relays is what causes the bulk of data usage, no?

If so, making clients more intelligent at deduping by using NIP-65/Gossip should be a **big** priority. Casual users aren’t going to be cool with burning through 30 GB of data and their battery life.
Avatar
hodlbod 2y ago

It's definitely relay redundancy, adding a proxy server to Coracle reduced bandwidth used by the protocol itself (not media) by 90%.

You can enable this in Coracle by adding `wss://multiplextr.coracle.social` to your "Multiplexer URL" setting (not to your regular relay settings).

nostr:nevent1qqsrgk7ygexg9nja2cz6lahdj6nrn5lc39wlnth02umv8zqvjegy0scpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsq3vamnwvaz7tmjv4kxz7fwdehhxarj9e3xzmny92m2uy

Reply to this note

Please Login to reply.

Discussion

Avatar
Andrew 2y ago

As far as I understand from NIP-65/Gossip, rather than pulling in everything from every relay a user is connected to, it'll pull in only what their follows have posted, from the relays they're known to have posted them to, right?

If so, then when we get to a point where most major clients use NIP-65, will multiplexer proxies no longer be needed since the redundancy is avoided?