figuring out exactly how far your posts is reaching is what my https://nrcheck.tigerville.no tool does. i don't publish to all the relays that are listed there for my posts, but i might still have followers who use them.

(oops, accidentally reposted that instead of replying)

one relay wouldn't necessarily know about all your followers. and you might even have followers on relays that you're not directly connected to.

that's been my life for the past few years now. when i do find stuff i'm motivated to do, it always gets me excited.

assuming that "who follows npub X" is a query i can run on relays. will have to check docs...

it's not the easiest tool in the world to make, but i *think* it's doable.

500000 sats is worth about an hour of someone's time, assuming they get $20/hr. 1600-1700 sats is worth about a minute.

the world average wage is roughly 10 satoshis per second or 583 per minute.

that's to the point where more life energy was wasted checking who zapped you than what the zap was worth

i use crypto in the same way polytheistic religions just add new gods when they encounter other religions, instead of abandoning all their old beliefs. like the Japanese kept shinto even as they adopted buddhism. i'm no investor so i'm mainly concerned with where the doohickeys are coming from and who accepts doohickeys as payment and adapt accordingly. 🤷‍♂️

i thought satoshis were quite an optimistic unit as for the expected deflation of Bitcoin but millisatoshis is *really* optimistic

TIL millisatoshis are a thing

there's just one challenge that i was thinking about earlier: could it be in the interest of a relay to fake being paid?

i've been working on the https://nrcheck.tigerville.no Nostr relay checker since Wednesday, slowly improving reliability and adding features, and more is coming.

i better figure out how to check if relays are paid or not from JavaScript, if there is even a standard way of doing that...

okay, i think the code for NRCheck is slightly less ugly now. doesn't hurt to clean it up. in the parts of the code where i have to talk to the callback API of nostr-tools, i'm using promises, and in the parts where that's not needed, i'm using async/await, since you can await any function that returns a promise, not only one declared as "async" (it's just syntactic sugar).

been touching up https://nrcheck.tigerville.no more. not much in terms of visual changes, but it's a little bit cleaner now. about to reformat the result lists as HTML tables since it's tidier to have the latency times in a separate column.

(oops, accidental repost, meant to click reply)

front room? no idea what you mean. the clue is in which country spells it like Nikolaj and which country that country is near to...

tbh i wasn't even thinking about fakery when i made this. do keep in mind that paid relays will also not relay your posts, so they aren't "faking" it as such. might split out the paid and free relays in the results lists, now that you mentioned that, though

client filtering would be helpful. there isn't full agreement between clients and relays about who filters what yet. iris.to wants to do all the filtering by keeping a big social graph database in the client. meanwhile, filter.nostr.wine is basically that except the relay does it

for you...

if i'm going to venture a guess as to why some relays connect but don't respond to queries, it's because they're special-purpose relays that aren't storing any events. another guess would be that they don't like it when the first query is for a note or event. come to think of it, i'm not handling the case where they just disconnect you immediately after the query was sent.