@arthurfranca
Software dev + graphic designer + Nostr enthusiast
https://github.com/arthurfranca
Relays (2)
- wss://relay.44billion.net β read & write
- wss://nos.lol β read & write
Recent Notes
Two corrections:
1. "Gateways can inject whatever they want into nsites, and by extension napps." This was erroneous and obfuscated, 44billion originally served "napps" but the "napps" ontology was c...
Anything I say will sound like I'm protecting 44b.
I'll just add that napplets running both on web and on native platforms is what would make napplets reasonable. Though when napplets use html and css, it forces the native shell to embed a full browser engine, thus not being really native anymore. Thus to me napplets currently are nsites with arbitrary constraints.
I didn't look too closely at Verdana napps, but I think it started walking the same path of napplets in this regard, but now it is moving to wasm and that would make much more sense. Although it may initially suffer from less adoption cause html and css is easier for devs, the profession is dying and everything is easy for AI.
I can switch or add support to the winning spec anytime. I just don't see a reason at the moment.
I'll just add that napplets running both on web and on native platforms is what would make napplets reasonable. Though when napplets use html and css, it forces the native shell to embed a full browser engine, thus not being really native anymore. Thus to me napplets currently are nsites with arbitrary constraints.
I didn't look too closely at Verdana napps, but I think it started walking the same path of napplets in this regard, but now it is moving to wasm and that would make much more sense. Although it may initially suffer from less adoption cause html and css is easier for devs, the profession is dying and everything is easy for AI.
I can switch or add support to the winning spec anytime. I just don't see a reason at the moment.
1
Why is it a tough sell? There's 500 napplets and more than 10 shells. Anyone produce functional, bug-free and secure napplets in less than 5 minutes. My record is 72 seconds. nostr:npub1nmr6w7qk0ta36...
I mean, take a website, remove all its html files and leave just the index.html one, remove all web apis like bluetooth and fetch and you get a napplet. Then reimplement these web apis you previously removed access to inside a parent website so to make them available to the napplet. Like... why?
Before you pivoted it to nostr **web** applets it made more sense.
Before you pivoted it to nostr **web** applets it made more sense.
1
does Verdana work with napplets or the other napps spec nostr:npub180cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsyjh6w6? i'm confused.
nostr:npub1l3cgtsurhfchg4cyhhqudm70074sr96srhje330xc5m6cze...
44billion only supports nsites. I introduced some optional `window.nostr` and `window.napp` apis.
Napplets are like nsites with no access to browser apis - tough sell. Only SovEng pals are building napplets. They use `window.napplet` api. Verdana introduced many competing `window.napp` apis.
Three different event kinds. Pick your poison.
Napplets are like nsites with no access to browser apis - tough sell. Only SovEng pals are building napplets. They use `window.napplet` api. Verdana introduced many competing `window.napp` apis.
Three different event kinds. Pick your poison.
1
Speed up. Bugs dowm. https://zillion.chat/
https://blossom.primal.net/e372aa5a991a007f5bafc32dddeef10439874365754adfb2b5f73c9e63c855e4.webm
https://blossom.primal.net/e372aa5a991a007f5bafc32dddeef10439874365754adfb2b5f73c9e63c855e4.webm
2
whoa dude what is this?
You work reminded me of this Sega Genesis game https://en.wikipedia.org/wiki/Shadowrun_(1994_video_game)#The_Matrix
π€©1
What protocol does it use?
I'm calling it p2r2p. It isn't only for chat apps, it's for any private data exchange.
1
Updated our website to get rid of all the BS: https://amethyst.social/
Straight to the point. Very good!
π1
Nostream 3.3.0 supports the proposed NIP: https://github.com/cameri/nostream/releases/tag/v3.3.0
Let me know if weβll need to make changes
Nice! I'm about to add support for it too on my relay.
I see you added the `params` value to `OK` messages but not to the `CLOSED` ones yet.
I see you added the `params` value to `OK` messages but not to the `CLOSED` ones yet.
1
I also noticed that relays in general have no way to inform for how long should clients backoff.
I've just submitted a PR to address that problem: https://github.com/nostr-protocol/nips/pull/2498
Pi...
Thanks for the feedback, all of you. Edited the PR.
I see that @Vitor Pamplona and @jex0 seem to prefer miliseconds granurality. I'll change that too.
I see that @Vitor Pamplona and @jex0 seem to prefer miliseconds granurality. I'll change that too.
2
