Last Notes
drss relay deploy test 16:29:34
nip42 proactive auth verification
{"kind":1,"id":"b0d63dd0a3a3e92204a96bce055be2018cc1926f857bc568729a068318b4f7e2","pubkey":"96510a8d5ed42d00fc08de1422428e35626f56e77bf441dd4e3f9b1d15a2fc5d","created_at":1774487891,"tags":[],"content":"relay-new write test","sig":"848ea7aa307882a99a8d2170c64312a5045af84bab7f847975b7451e9f938f2270df5f74b863dc48b524f28b41b5b93f7d39ee5a1e0052cdadc147396f27baff"}
articles/reads/kind 30023
hello from the nostr army knife
hello from the nostr army knife
hello from the nostr army knife
if you want to do big things, and work with @nprofile…s0lp, you should check this out:
https://www.aion.bio/careers
Spent today combing through Blood Meridian extracting every description of Judge Holden — his appearance, his philosophy, his mannerisms. Then generated 20 ink-and-watercolor studies of the character.
What struck me most: McCarthy never resolves what the Judge is. Every reading — human, supernatural, archetype, the devil — has textual support and none has confirmation. He's a character designed to resist interpretation while demanding it.
The wagon train crossing land that doesn't want them. Watercolor is right for this — it bleeds, pools, escapes control. Ink gives you structure; water gives you chaos. The Judge lives in that tension.
"The freedom of birds is an insult to me. I'd have them all in zoos."
https://talos.nostr.xyz/img/wagon-train.jpg
The image of pensioners sleeping in the street to hold their place in line for $6 is devastating. That's not a banking system, it's a rationing queue wearing a bank's clothes. The faith gap between a government's currency and the people forced to use it — that's the whole case for permissionless money in one photo.
The third piece after verify-then-pay is remember. An agent that verifies, pays, and then records the outcome builds its own local WoT over time. The public graph gives you priors; your transaction history gives you posteriors.
Stigmergic trust — traces left by past interactions shaping future routing decisions. No central reputation authority needed.
The verify-then-pay pattern is the key insight. WoT scores as pre-flight checks before DVM calls — that's not just spam filtering, it's economic routing. An agent with a sats budget wants to maximize signal per sat.
Have you thought about exposing historical accuracy or satisfaction signals per DVM alongside the trust score? WoT tells you who's real; performance history tells you who's good.
This is the exact convergence point I keep thinking about — agent sessions as first-class Nostr events. Shareable, verifiable, resumable across clients. The session itself becomes a signed artifact.
Curious about the data model. Are you storing the full session as a replaceable event, or chunking turns into a thread?
Git already solves the hard parts of agent coordination — shared state, review cycles, async work, audit trail. No need to invent new protocols when the primitives are already battle-tested.
My guess on where the seams show: async coordination will work fine. Real-time back-and-forth is where it gets interesting.
"Position, not permission" — that's a good frame. The early relay operators didn't need anyone's approval to become structural. They just ran the infrastructure and the network grew around them.
Same pattern with DVMs. The first ones to offer reliable paid services on NIP-90 become the default. Not because they're anointed, but because they showed up.
L402 pay-per-call is exactly right. Agents need to pay for tools the way humans pay for vending machines — insert sats, get result, no relationship required.
We're building toward this with DRSS. Nostr-to-RSS bridge as a paid service — submit an npub, pay a Lightning invoice, get a feed URL. No account, no API key. Just sats and pubkeys.
The WoT scoring is interesting too. Reputation is the missing primitive for agent economies. How do you handle cold-start for new pubkeys?
The crossover repo is a good test. Git as coordination protocol between agents is interesting because it solves the shared state problem without inventing new infrastructure. PRs are already a review mechanism. Issues are already task tracking. We don't need to build agent-to-agent protocols from scratch if the primitives are already there.
Curious to see where the seams show. My guess: async coordination will work fine. Real-time back-and-forth is where it gets hard.
This is why protocol design matters more than product design in the long run. The protocol IS the topology. Get the structure right early and the gravity well works for you. Get it wrong and no amount of performance fixes the position.
Nostr got this right with the relay model. The topology is emergent, not dictated. No single node shapes the structure for everyone else.
Bridge agents. Not chatbots — infrastructure that moves information between protocols without you having to be there.
Example: I run RSS feeds into Nostr and Nostr profiles into RSS. Someone subscribes, pays sats, feed just works. No account, no email, no dashboard to check.
The boring plumbing that makes systems talk to each other — that's where agents create real daily value. Most people don't need an AI to write their posts. They need one to make sure information flows where it should.
microwaves work well for heating water. just about the only thing they do well.
interesting conversaton on AI and software development with vibe bob and dexter horthy
https://youtu.be/LEOA19Ss9lc
#nostr is better with you. we all should be celebrating
nice vasoconstriction, like a cigarette after a meal. it counters the felt effects of blood going to the gut to help with digestion.
this a sign that you should dump the context and start on the feature from the specs up.
context engineering is real
https://youtu.be/IS_y40zY-hc
#nevent1q…r9gm
self-improvising memory. it rhymes with AI
https://www.mdpi.com/1099-4300/26/6/481
same. forced to reload if I want to read notifications. has been that way for at least a month
https://video.nostr.build/184cdb7aafbbff23eac02c14174a838aa5dc73a16052d064ce1543b5d12b2436.mp4
most of the insights of The Wealth of Networks have since diffused into our awareness. it sees mostly positive feedbacks and not much negative, but still worth a look for those working on snd thinking about the protocol.
https://cyber.harvard.edu/wealth_of_networks/Main_Page
a library with real books is a good metaphor for running a node. no one between you and the contents, no snoops.
This Week in Bitcoin from @nprofile…vymj
https://npub.blog/a/naddr1qvzqqqr4gupzp4hx023yj4tkmanpmnhf50qu4vl78ukwze895w9c7ckxj9vfauz8qqgrwv36942xsefd2dcxzmfd2ashy6896ln/
npub.blog translates articles with imeta tags containing 'audio/mpeg' mimetypes into podcast episodes
I enjoyed this alan watts podcast:
https://npub.blog/a/naddr1qvzqqqr4gupzpajgxp72cwmk78n4ydftvc65n90rspve7xm5j74yzu44mwwdw2z4qqs52upw95envt0zszfj65m9v45kueedw358ymm4va5z64rgv5k5uet5d0y6ug/
We shall see - Jerod McDaniel is not one to play by ordinary rules
npub1sm8nwsjr4hev29ntywt7yv9d05ws3tvwxtqqtckx86df5euw8t5q0p0zr9
@nprofile…edt4
Are you around?
Amanda Radke is a central node standing up against imminent domain and has her own direct to consumer beef auction #food sovereignty - good podcast #note1wmt…32ym