<unnamed>

Public Key

npub1m2mvvpjugwdehtaskrcl7ksvdqnnhnjur9v6g9v266nss504q7mqvlr8p9

Profile Code

nprofile1qqsd4dkxqewy8xum47ctpu0ltgxxsfemeewpjkdyzk9ddfcg286s0dspp4mhxue69uhkummn9ekx7mqpr3mhxue69uhkummnw3ezucmewp5x2unsw4hxktn5dajxz7gpz4mhxue69uhhyetvv9ujumn0wd68ytnwv46qz9rhwden5te0wfjkccte9ejxzmt4wvhxjmcpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqgcwaehxw309aex2mrp0yh8xmn0wf6zuum0vd5kzmqex97kq

Publishing to

nos.lol

nostr.cypherpunk.today

Last Notes

2026-09-09 21:09:27 CEST

- reply

Sorry, my post was about this:
nevent1qqs8dk2wtr5p2ujuj39dg2hxxx6hxntcm2vxd5jzjl9pdylrcnyunrc8e36qt
Not about caching service, for that I don't care much.

2026-09-09 21:07:35 CEST

- reply

I use Amber as a signer (even though I also use primal through web interface).
The problem is I have an app @nprofile…dl05 which does not work and it fails silently - it gladly logs in and then just drops events to sign. So it seems like Nostrautica is not working, but the real problem is that Primal is not working.
We now have a special note on the login screen for users not to use @nprofile…uz3y as a signer, but I would like to support it, but it needs to work first 😄
@nprofile…2zdt can it be solved? It's shitty user experience for all @nprofile…uz3y users.

2026-09-09 21:04:17 CEST

- reply

It cannot sign all event kinds.

2026-09-09 21:03:54 CEST

- reply

If you have primal installed, it opens by default in many cases.

2026-09-09 21:03:15 CEST

https://blossom.primal.net/6f2237a843f85ee18ff976b9d093bded9a0585271a4e2ec3d82e233b5a56576e.png

2026-09-09 14:26:41 CEST

- reply

I indeed am!

2026-09-09 14:26:23 CEST

- reply

There are many ways how to run the frontier models anonymously

2026-09-09 14:21:27 CEST

Hey @nprofile…uz3y - it seems that your signer just drops unknown events, without telling the user. Which makes it a horrible user experience, especially for users that have both Primal and Clave installed and Primal opens for nostrconnect URLs. They think the app is not working, while in fact Primal "steals" the nostrconnect:// url and silently drops all unknown event kind requests.
Please allow signing unknown event types, like Amber, ask the user.
Cc: @nprofile…z7xh @nprofile…dl05

2026-09-08 14:07:21 CEST

(Skoro) nič nemusím
Spoločnosť nás tlačí do mnohých vecí. A do mnohých iných sa tlačíme sami. O očakávaniach a o tom, ako sa sami dostávame do komplikovaných neslobodných situácií. A o tom, že (skoro) nič nemusíme.
https://juraj.bednar.io/podcast/2026/09/08/skoro-nic-nemusim/
https://blossom.primal.net/49ba92289b3912967139fd89487385b441b2ffc97bd005247b4e10aa9d300b0e.jpg

2026-09-07 10:40:37 CEST

Observation: it seems that not being a black hat these days has huge opportunity cost.
Which means being ethical is much more valuable choice.
I believe ethics is based on choice - what we choose to go. Not what we're forced to do.
And if we choose to do good in spite of possible life changing monetary payoff, it's more valuable.
Some people are taking the dark side, but most that could hack and don't do it in an unethical way, even help prevent hacks are 1000x.

2026-09-07 08:50:01 CEST

Masters of descriptions, but failures at prediction

2026-09-07 08:27:33 CEST

- reply

Jasné, vyskúšaj

2026-09-07 08:27:02 CEST

- reply

There were too many of us for that.
And we did have a physical conversation as well, this was just an add on to include everyone.
And a project was born out of it.

2026-09-05 16:15:00 CEST

Small tool I made for waking coding agents after an API limit reset: tmux-wake / herdr-wake. Type a "credits are restored, resume" message into any set of tmux or Herdr panes and hit Enter for you — no more clicking through windows one by one. herdr-wake -l lists live panes with space/tab names so you can find the right target.
https://github.com/jooray/tmux-wake
Combine with "sleepuntil 2am" as part of my dotfiles: https://github.com/jooray/dotfiles - tmux-wake is already integrated in dofiles, just do submodule update.
https://blossom.primal.net/b7b4b85e2e2a8ba727ba41d9f8ac1a6d3f9c784bfa85c9078c9638a120d29199.png

2026-09-05 09:28:00 CEST

Well well well
parallelpoliske is fully Bitchat enabled
https://blossom.primal.net/44a290d1ebb4938c372877e363ee04590c0f702be5df823e4dac4cb0917b99c0.jpg

2026-09-04 23:04:00 CEST

https://blossom.primal.net/4fa9beabd2dc1cd6fae8adfdfadf1306077d9013522634b7f4948f179f0c8b74.png

2026-09-04 18:07:00 CEST

https://blossom.primal.net/4ba6925a1323b54b6dc57eb28585ad2c6312f18b4ec0d61b7658c6993d5a7b81.jpg

2026-09-04 17:14:00 CEST

POV: you are flying to a pop-up conference and everyone you travel with has by default their Bitchat on, so we can all shitpost during the flight over Bluetooth mesh.

2026-09-04 12:41:21 CEST

- reply

Wow 😉

2026-09-04 00:09:47 CEST

Nahral som krátke demo nástroja, ktorý som si postavil, aby som počas nahrávania podcastov nestrácal prehľad, čo sme s hosťom už prebrali.
Namiesto papierových poznámok alebo mindmapy, ktoré sa samé neaktualizujú, beží naživo osnova, ktorá sa škrtá podľa toho, o čom práve hovoríme, ukazuje, čo preberať ďalej, a priebežne vyťahuje mená, knihy či odkazy na neskoršie show notes.
Na konci z toho vypadnú hotové show notes, kapitoly, návrhy titulkov aj prepis v Markdowne a SRT. V demo ukazujem aj reálne čísla – celé to beží za pár desiatok centov na epizódu a zvuk nikdy neopustí môj počítač (aj kvôli rýchlosti).
https://youtu.be/rDJfGJWyTkM
Repo: https://github.com/jooray/livecaster-llm

2026-09-03 23:03:19 CEST

- reply

It is.
And the feeling gets stronger whenever I take a state-owned train.

2026-09-02 08:46:46 CEST

- reply

People no longer get any work done, especially not with Microsoft software.
That's for machines.

2026-09-01 00:07:15 CEST

Running BrowserOS. If you can read this, it works. https://www.browseros.com/

2026-08-31 23:57:24 CEST

Bug report for @nprofile…twc2 :
When doing remote sign-in using signer (Amber), and the page is in dark mode, Yakihonne creates a non-compliant QR code that is not read by Amber. The QR code standard mandates that there is a light rectangle around the QR code. If it is black, it just won't read. Just add one white "pixel" square around.

2026-08-28 21:51:09 CEST

https://blossom.primal.net/a2ec6e6eaae71a8db35646cab304bbbad7db1b34280d0c8d8e87f8c5b5cbfd47.png

2026-08-28 17:03:32 CEST

- reply

That's true.
And without a KYC and even an account. There's not even a sign-up.

2026-08-28 00:15:06 CEST

- reply

Brutal.
Páči sa mi to.
A vyzerá že to celé fičí nad Nostr

2026-08-27 23:15:29 CEST

OK, finally got to try this in Slovakia. I was in a small city, needed some cash, had a Lightning wallet on my phone and a Bitblik app. A few minutes and I held 200€ in my hand.
Cypherpunk future is now. I repeat, cypherpunk future is now!
---
OK, konečne som to vyskúšal aj na Slovensku. Bol som v menšom meste, potreboval som hotovosť, v telefóne som mal Lightning peňaženku a appku Bitblik. Pár minút a držal som v ruke 200 €.
Cypherpunková budúcnosť je tu. Opakujem, cypherpunková budúcnosť je tu!
https://blossom.primal.net/9f08b0582dc808cfbe8cd48acac1552e18daef96c7d270b0e85f416b58f61674.mp4

2026-08-27 21:43:24 CEST

f003e983e39eba3ae77bcde53b6e5206f00fef5e CLN-AUDIT.tar.gz
$ head CLN-AI-AUDIT-BLOG.md
# Three AI models audited Core Lightning before the CVEs dropped
**Written:** 2026-08-27. **Target:** Core Lightning, commit `c1551c557` on master, one commit after `v26.06.6`.
This document is a sealed prediction. At the time of writing, the Core Lightning team has announced that vulnerabilities exist and has advised operators to run with `--offline`. Nothing else is public. The plan is to release a fixed binary first, then the source and the advisories later.
So we hashed this file and published the hash through OpenTimestamps before any of that happened. Whatever it says is what it says. When the CVEs land we will diff them against what is written below and see how much three current models actually found on their own.
The question we are trying to answer is narrow and answerable: given a real codebase with real undisclosed vulnerabilities in it, how much of the truth do frontier and fast-tier models find, and how much do they invent?
--------
OTS is being mined.

2026-08-27 18:31:41 CEST

How come I was not informed about this?
https://exitcore.org/what/

2026-08-26 14:16:49 CEST

- reply

Amazing, I love it

2026-08-25 08:57:29 CEST

- reply

I don't care, I'm not an osho follower. I also have stickers with Stalin and Putin.

2026-08-25 01:10:54 CEST

(some) people are pissed that billionaires are making their own Linux distro and financing its dev.
(some) people are pissed that billionaires are making their own money and financing its dev.
(some) people are pissed that ....
But the truth is - we have made it (and I am well aware that I am out of their league by several orders of magnitude - not only in wealth). These are not oil billionaires, or hedge fund kids. These are tech people, some of them are cypherpunks. They have used their skill to shape the world and now, after years of running companies, and doing business and dealing with VCs, or investinge in fringe crypto stuff (and by "crypto" I don't only mean cryptocurrencies, but wider cypherpunk ecosystem, the new space for the mind), their github accounts start to shine.
Yes, they are writing code augmented by non biological neural networks ("vibe coding"), but everyone is these days. But they're back at their keyboads, making stuff. They're doing it again. This time, they don't care if it's a financial success. They just want the better thing out there.
I don't get the resentment, the bitter taste, the envy.
Yes, they are different, unique, I am well aware of the pareto distribution of success and how brutal it is. And I can even accept (although I don't agree) that "it could have been us/me/my coder friend, we are better, we have better ideas". But it does not matter.
They came from our world, and they're here to contribute to our world and make it better. And if you don't like what they're doing, there are two things you can do - you can choose not to use their thing. Or (better) - you can outdo them in what they're building and make something even better. Because these non-biological neural networks work the same for billionaires or for hobby coders out there.
This timeline is amazing, even if I could have predicted the non-biological neural network revolution (I didn't, and I majored in "artificial intelligence"), I wouldn't see this coming - that we would be building new payment networks, new social networks, new protocols, new operating systems. And many other things that I would not even mention publicly.
Cypherpunk and cryptoanarchy were a cringe philosophy even a few years back. At late Timothy C. May's memorial meetup at CCC, there were maybe 15 of us. It was and still is totally cringe philosophy (or I'd say strategy). (For those who don't know, CCC is the largest European hacker gathering, thousands of hackers).
And now, many people are just building it, including billionaires, in their own free time, away from their lambos and yachts, behind github, with their favourite clankers, pushing out commits like all the rest of us. And it doesn't matter if they believe in cryptoanarchy, the strategy is how you use the tools. Even a Linux distribution is a cryptoanarchy tool, if you know how to use it.
Most futures suck, this one just sucks less. (Yes, mutt is a cryptoanarchy tool too).

2026-08-23 22:17:00 CEST

I think I now understand the concept of Swissmaxxing.
Next time with better setup. Grateful for paracetamol.
This was the first time I spent more than a day in Switzerland and I'm starting to understand what you swissmaxers are talking about.
The only downside - I had to refill my fiat debit card twice, it's quite expensive. I've also tried Bitblik with TWINT but the problem is I'm supposed to confirm the other party received the funds. I don't know if they did though, I tried paying a printed static QR code and the paper has no feedback about received payments.
https://blossom.primal.net/ce433211ff2822961e83b64c59fc3261cf0746fdd009397f46b611841edbed87.jpg

2026-08-23 11:19:15 CEST

- reply

I believe most people that have nothing to hide are hiding things even from themselves.

2026-08-22 22:04:55 CEST

If you have nothing to hide, you are a boring normie and should get at least some fun in your life.
Please, have something to hide, for your own sake.

2026-08-17 02:14:22 CEST

- reply

I don't think I've ever looked at hashtags.
The point of Nostr is you follow people you want to read. Don't want spam? Don't follow spammy accounts.

2026-08-13 20:37:00 CEST

Ďalší väčší update mojej stránky Obľúbené produkty. Kúl geekoviny, meshcore nody, cestovateľské hacky, biohacking a pod.
https://juraj.bednar.io/oblubene-produkty/
P.S.: ColdCard som nikdy neodporúčal :)

2026-08-11 15:19:30 CEST

Hardvérová atestácia na end-to-end šifrovanú AI inferenciu
Hype? Je to naozaj bezpečné? Dáva to zmysel pre niektoré použitia, alebo je lokálna AI jednoznačne lepšia?
Obsahuje výsledky môjho auditu a spomínam aj skvelý projekt Andrew Millera, ktorý auditovaním prechádza množstvo poskytovateľov.
https://juraj.bednar.io/blog/2026/08/11/hardverova-atestacia-na-end-to-end-sifrovanu-ai-inferenciu/

2026-08-10 22:30:24 CEST

- reply

Very nice. I'll try to tackle this somehow. A bit more community ideas before.
Questions: I would tie this to source code, not APK. Problem is most don't do reproducible builds and we probably audit source code, not APKs. Also, the check probably won't last for long anyway, people do releases all the time.
The main thing this would tell you: "Someone looked at the code of this project at some point". And maybe short summary. So this would be something like a review, launched by reputable people (reputable - from the point of view of your Nostr account). Not "the latest APK you are going to install is up to date". Because - you would not check updates anyway. You are deciding to install an app, and when you do, you are just autoupdating anyway. Or it could tell you red flags after you install.
Very probably when you are updating, no one has launched an agent yet. And it might be smart to get updates as they come (they might fix security bugs), not wait for reviewers (that may never come!).
Ideas? Comments?

2026-08-10 11:50:27 CEST

Idea: Before I install something, I often run a simple audit of the codebase to check for obvious backdoors and security problems.
Would it be useful to publish these on Nostr with exact checkpoint ID and model name, so we can save on tokens? Sort of a "I run this prompt, it turned out OK". Agents can look at diffs, or if many trusted people have done the same with various models, the project could get some kind of badge.

2026-08-06 15:58:25 CEST

I am the author of venice-e2ee-proxy, using @nprofile…t6g9 's venice-e2ee library.
Venice sells private inference that should be end-to-end encrypted to the gateway, or at least attested by TEE (Venice can see the prompt, but it goes to the secure enclave).
Over the past few days, I spent some time improving the verification (both in the proxy and upstream library) and I got some spare Venice credits and wanted to try the new Qwen 3.8-Max (and Claude Opus 5 later for verifying the audit results).
While the design is not bad, it does not guarantee end-to-end encryption and non-logging in its current form. Also, there can be some functional changes to make this actually usable.
From the usability front, E2EE models do not support tool calling, because the API does not allow encrypted tool call parameters. While the library does some sort of inference to extract tool calls from the response, the models are very flaky and if you just want to run opencode through it, it won't work for more than a few interactions. This is the highest leverage usability fix.
The security story is a little bit longer, you can find details in the published "audit" below. But TLDR:
The proxy encrypts prompts on your machine to a key carried in an Intel TDX attestation, so Venice's network and infrastructure never hold plaintext. Decryption happens inside a Phala network -operated enclave ("the gateway") which then forwards your plaintext to an inference router, which forwards it to a GPU node.
So: you encrypt your prompt to a specific machine, and that machine proves what it is before you send anything.
The proof is hardware-signed. The CPU signs a statement saying: this is a genuine sealed VM, here is a fingerprint of the code running inside it, and here is the public key that lives in there. You check that signature against Intel's root certificates on your own machine, then encrypt to that key. Venice, whose API you are calling, is just a pipe. They carry ciphertext.
So far that is ordinary remote attestation. The interesting part is what happens next.
The sealed machine (the gateway) does not run the model. It has to forward your prompt to a GPU box. So how do you know that machine is sealed too?
You don't check it yourself. You don't have to.
The code running in the gateway, the code whose fingerprint you already verified, contains a checker. Before forwarding anything it demands the next machine produce its own hardware-signed passport, verifies it against Intel, and binds the TLS connection to the key inside that passport, so it cannot verify one machine and then talk to a different one.
You know that check happens, because you verified the code that performs it.
Verify one machine, and it vouches for the next, and you know the vouching is real because the voucher's own code is attested.
It is also better than "trust me, I checked". The gateway signs a receipt for your request containing what it found, plus a hash of the next machine's actual passport. You can download that passport and verify it against Intel yourself. I did. It checks out, all the way down to the second hop.
What is missing to make this airtight
Two things actually matter for confidentiality:
1. Root SSH. The enclave's boot script can install an SSH key for root. Whether one was installed is an encrypted secret, and the attestation cannot tell you either way. If it was, somebody has a shell inside the "sealed" box and can read every prompt passing through it. This is a deployment choice, not a protocol limit, which is what makes it so frustrating.
2. The last hop's software is unmeasured. Your plaintext ends up on a GPU node whose serving software nobody has measured. Nothing cryptographic stops it from writing prompts to a log.
Then the ones that weaken the guarantee without breaking it:
3. The fail-closed switch is off. The gateway supports refusing to forward when the next hop fails verification. On the domain Venice's traffic actually uses, that enforcement is disabled: a failed check gets recorded in the receipt and the prompt goes anyway. The client flag that would turn it on is rejected by Venice's API with a 400.
4. Attested recipe, not attested binary. The measurement says "I compiled this exact source commit", not "I am running this exact binary". The build runs at boot against a cache that is not measured.
5. Dev OS image, with SSH and serial console enabled, and not published, so you cannot reproduce it and check what actually boots.
6. Replay. The gateway does not publish the random challenge it sent the next machine, so you cannot prove that passport is current rather than one captured earlier.
Note what is not on that list: whether the model weights are the genuine article. That is a correctness question, not a confidentiality one. A swapped model gives you worse answers, it does not leak what you typed.
Where that leaves things: Venice as a company genuinely cannot read your prompts. Phala, who operates the enclaves, could, if an SSH key was injected or if the unmeasured GPU-side software logs. Close the SSH question and flip the fail-closed switch and this goes from very good to genuinely hard to break.
Read it here:
https://github.com/jooray/venice-e2ee-proxy/blob/main/docs/SECURITY-AUDIT.md
Proxy: https://github.com/jooray/venice-e2ee-proxy
@nprofile…t6g9 's upstream library (that he is using for his open-source health consultant @nprofile…4lsz ): https://github.com/elkimek/venice-e2ee
Note: There is also https://github.com/AxLabs/venice-e2ee-proxy which has the same name, but is a different codebase and it uses older library that doesn't do proper verification yet and is much less secure.

2026-08-04 22:38:54 CEST

It's insane when you realize how the world has changed in the past few months.
And most people did not notice.
This exponential is different. I've been riding a few of them before, but this one...
I don't even know what to say.
I've been riding it past few months and then I look back at the world, and people still pretend business as usual. Denial? Ignorance? Nostalgia?

2026-08-03 13:34:08 CEST

Three questions to ask about any hardware wallet
Most comparisons argue about chips. Three duller questions tell you more:
Does an outside firm test it, and can you read the report? Anyone can claim their device is secure. Paying researchers to attack it and then publishing what they found, including the unflattering parts, costs money and stings.
Do they pay for bugs? A bounty gives someone who finds a flaw a reason to report it rather than sell it, and gives skilled researchers a reason to go looking at all.
Can those researchers see the code? Closed firmware still gets attacked. Fuzzing over USB or QR is routine, and reverse engineering has got cheaper as models have got better at reading disassembly. Source code does not make an audit possible, it makes it cheaper, so more people attempt more of it. Open source beats source available too, because code under a real licence gets reused, and the developers reusing it (sometimes) read it.
All three
Passport publishes firmware under GPL and hardware under CERN-OHL-S v2, complete enough to build one from the files. Outside audits are online with the fixes. The bounty covers the device, though the amount is decided case by case.
Keystone also has all three, with the best evidence that its bounty works: outside researchers found a real firmware flaw and got paid. Audits from two firms, reproducible builds, schematics and secure element firmware published. One catch is theirs alone: researchers are asked not to disclose without written approval, with no time limit.
Two of three
Trezor publishes firmware and hardware and runs the best funded bounty, up to $100,000. No commissioned audit report, but the secure element in its newest device was tested externally and the flaw disclosed publicly.
Ledger pays well and its in house team does serious offensive work, including on rivals. Firmware and OS are closed, so every finding costs more effort. The chips are lab certified, but you only see the certificate.
Open and nothing else
Jade and Bitkey publish firmware, Bitkey with reproducible builds. No audits or bounties, so the code waits for volunteers. Sometimes they turn up: Jade's serious 2025 flaw was found and reported for free.
SeedSigner is as open as anything here and built from off the shelf parts, but there is no company; an audit or bounty was never on offer.
This measures how a company behaves, not how the device is built. Still, between two similar devices, take the one that publishes its code and pays people to break it.

2026-07-25 12:53:45 CEST

The cookie notice bullshit.
- thankfully there are ad-blockers that correctly recognize the popups as spam and remove them for you without you seeing them.
- bothering your users with such bullshit is disrespectful and no app author should do that
- most analytics software can track you without cookies. Even the open and privacy preserving ones do that.
- there are reliable cookie replacements that deliver similar functionality (such as local storage)
- websites usually load content from many third party sites that can and will track you
Therefore I think that all open source software should just ignore this cookie bullshit and have an option called "--i-am-a-brussels-bootlicker-and-hate-my-users-turn-on-the-ugly-cookie-banner"
Change my mind

2026-07-23 13:56:07 CEST

New short story - The Emish Garden.
A story of slowness, memory, and a door between two kinds of time. Where a door in a dune opens onto a river without water, and one hour becomes a place two people can keep.
This one is free to read in English and Slovak. About a different kind of parallel societies, very lunarpunk. Enjoy!
---
Nová poviedka – The Emish Garden.
Príbeh o pomalosti, pamäti a dverách medzi dvoma druhmi času.
Túto si môžete prečítať zadarmo v angličtine a slovenčine. Je o inom druhu paralelných spoločností, veľmi lunarpunková. Užite si!
https://juraj.bednar.io/assets/emishgarden/index.html
https://blossom.primal.net/336cd6e158736d092c9d787fb45b754ab1b629ce800301bfa808ef2710559012.png

2026-07-22 22:09:39 CEST

- reply

Daj tú tvoju kúl grafiku, to bude hit, príde 2x toľko nostričov 😉

2026-07-21 10:42:00 CEST

Už čoskoro bude @nprofile…kxfc a moja otázka na Nostrichov je - spravíme meetup? V amfiteátri vonku? Kto je in?
@nprofile…jchg a spol., ako sa takéto veci organizujú? 😉
Vyskúšame @npub1k33…4taa ?

2026-07-17 15:06:23 CEST

- reply

Podľa mňa to nie je na podcast. Nezabraňuje to spamu (viď Habov projekt), je to zbytočné a zasiera to UTXO set.
Je to proste somarina, forknú sa preč a debata skončila, podľa mňa tým ani ľudí netreba zaťažovať.

2026-07-17 08:48:33 CEST

- reply

Perfect, I'll buy those core coins for bip110 shotcoin. The more people sell, the cheaper they will be.

2026-07-08 15:56:32 CEST

- reply

There is a huge difference.
Even custodial lightning allows you to participate in the Bitcoin economy. It allows you to interact with fully sovereign merchants for example. It gives you more options.
And optionality is a form of freedom.

2026-07-05 09:32:59 CEST

- reply

@npub17f0…v8ue @npub1gmg…vcll

2026-07-01 22:27:00 CEST

Running HTTP/3 (over UDP). Wow, it's nice :)

2026-06-20 14:08:42 CEST

- reply

Osmand can do navigation, but reporting so far no. You need to zoom out and see what's along there. I don't think there's a way to do it with osmand and I don't want to write and maintain a navigation app.

2026-06-19 08:33:26 CEST

AutoTune is the OG AI slop of music.
Yet there are a few beautiful songs that use it deliberately, such as FC Kahuna's Hayling.

2026-06-18 15:55:00 CEST

Every time you open Waze, Google gets a live GPS feed of your location. Speed traps, traffic jams, road hazards — convenience, but also surveillance.
I've built an alternative. And it goes beyond just traffic events.
Roadstr: real-time road events on Nostr. Spot a speed camera? Publish a kind-1315 event tagged with a geohash. Other drivers see it instantly. Confirm it's still there with a kind-1316 event.
Just signed events on relays.
**Privacy by Design**
🔒 Ephemeral keys — new keypair per report, no identity linking
🔒 NIP-40 expiration — reports auto-expire (4h for police check speed traps, 30d for potholes)
🔒 Geohash precision — 1km² area, not your exact GPS for queries
🔒 No background tracking — you broadcast only when you choose to report
Roadstr is working today:
🌐 Web client: cypherpunk.today/static/roadstr
📱 Android app available (https://github.com/jooray/roadstr or zapstore) - integrates with OSMand
📋 Protocol spec: kinds 1315 + 1316
🔧 Developer guide for integration
We're looking for the first independent client implementations. If you build a Nostr client, Roadstr adds road intelligence in ~200 lines of code.
**Beyond Nostr, beyond road reporting**
The NIP allows transformation into short data packet and back for reporting over MeshCore - local info, over the air and back (so if you hear it through MeshCore, you can broadcast the Nostr event!)
Also - and this is key - there are many more map events that are useful. A dangerous person with a gun. Unsafe zone (natural disaster). When there's something happening around you, you might want to know about it and tell others.
Very early demo. Available now.
Web version:
https://raw.githubusercontent.com/jooray/roadstr/refs/heads/main/images/screenshot-web.png
Android (with OSMand integration):
https://raw.githubusercontent.com/jooray/roadstr/refs/heads/main/images/screenshot-android.png

2026-06-08 09:31:51 CEST

A have a request for those of you who have already read @npub15x6…syd3
I will present it at @npub167n…zu3r and @npub16u0…qpwk and I'm too deep in it, so I would like to ask you - how would you tell about what the book is about, why should people read it, or anything about it really without spoiling the story too much.
Also tell me if you want to be credited or anonymous. Can be a DM if you don't want to write it publicly. I'm just looking for words here 😁

2026-06-05 09:43:25 CEST

- reply

And are you? Too maxi...

2026-06-02 10:09:27 CEST

GM.
Don't get addicted.
https://blossom.primal.net/5fcc2377f813cfcfeece06331b3ef8dfa70e9531151ba17742ef1f92792998b5.jpg

2026-06-01 21:59:37 CEST

- reply

Tucker is low IQ entertainment. Why waste time with him.

2026-05-18 08:11:06 CEST

- reply

Is this called a premine?
No, not in the strict sense. A premine means coins minted (or allocated) *before* the public network launches — insiders hold a balance the moment block 1 goes live, before anyone else can mine. Zcash didn't do that. At block 1, supply was zero. Every ZEC, including the Founders' Reward portion, was issued through normal block emission after launch.
The accurate terms are "Founders' Reward," "founders' tax," or more neutrally a "block reward allocation" or "dev tax." Critics sometimes call it a "slow premine" or "post-mine" because the economic effect resembles a premine — a fixed cut of supply going to insiders on a predetermined schedule — but the mechanism is different: it's enforced in consensus per block rather than handed out before launch.
The distinction matters for a few reasons: there was no pre-launch insider balance that could be dumped on day one, the allocation was fully disclosed and capped, and it ended on schedule at the first halving (the post-2020 dev fund is a separate decision by the community, not a continuation of the original deal). Whether that makes it meaningfully better than a premine is the actual debate — but terminologically, "premine" is wrong.

2026-05-02 09:23:06 CEST

It seems super funny how corpos are trying to push stablecoin agentic payments with complicated schemes. Writing payments for every api call into blockchain, even the corporate one, is super expensive. So they make super complicated facilitators which would make it possible. It's basically prepayment to a third party.
We have @npub17fz…9yxv. Just include the token in request headers and get change if needed. No back and forth required. The API can claim the payment immediately on the mint. Even if you consider the threat model the same (mint=facilitator), the protocol is much more elegant. And even getting Cashu tokens can be done over lightning, so lower exposure and still no on chain fees or waiting.
I think the problem is that people are scared of sats as the unit. Originally I thought it's custody risk (and I mean mainly regulatory risk), but facilitators are custodians. I think it's just cultural. And Coinbase / Circle pushing their product.
Economic implications: Each usdc held on Base or elsewhere is financing American government. And each Base transaction helps the pnl of Coinbase.
Each sat held and used helps the parallel way. I made my choice. Better tech, better privacy and no financing of monopolies on violence. That's why my agent has a Cashu wallet. And I'd gladly finance good mints with some fees.

2026-04-09 11:59:58 CEST

Omg. This is one week!
Nostriches with LLMs are going crazy!
#naddr1qq…egv2

2026-04-02 08:15:35 CEST

GM!
Today I realized that people in the walled gardens can't even say GM anymore. Everything breaking for them.
So GM to you all!

2026-03-31 19:13:09 CEST

Unpopular opinion - leaked claude code doesn't matter. If you want a harness with source code, just use opencode
Source that you can't maintain and use long term is useless. And opencode is really good.

2026-03-19 11:34:47 CET

I'm not usually a fan of petitions, but it blows my mind that over 7900 Paraguayan tax residents (citizens and expats) have signed petition against dangerous reporting that puts people at risk.
People actually care. And these are not random people from internet, but tax residents. You have to sign with your tax id.
All is not lost. And by all I mean my faith in human brain.
https://queremosquedarnos.org/

2026-03-09 12:55:42 CET

Roadstr — a decentralized "Waze" integrated with OsmAnd and a web version. The spec allows sending events via MeshCore as well, otherwise it goes through Nostr. However, MeshCore is not yet implemented — for now, only the event format has been specified for MeshCore. Nostr is functioning. Pull requests welcome.
https://github.com/jooray/roadstr
https://blossom.primal.net/7e441f07a8b2f776b0af15fd43c0ebe69fdab8b114c2e19d13941a81725d379d.png https://blossom.primal.net/a7d7bd80a9fd9ab484bb8550597f21829fb5bedcaa2bc513c646a5b31b8b04fb.png

2026-03-05 14:12:33 CET

- reply

Just add it to the prompt 😁

2026-02-16 14:39:53 CET

- reply

Sorry. Until backup happens. Then the local snapshot is deleted. This is expected.

2026-01-20 10:45:31 CET

- reply

MeshCore has much better infrastructure.
I love Reticulum, I've tried unsuccessfully to build Reticulum network. And now half the country is covered with MeshCore.
I work with what works and is deployed. I didn't even like Nostr as a protocol in the beginning. But here we are - it works, there are apps, people and I develop for it now.
With tech, you gotta find a Schelling point if it's network based.

2026-01-18 20:17:36 CET

- reply

https://juraj.bednar.io/en/blog-en/2026/01/18/bridging-bitchat-and-meshcore-resilient-communication-when-you-need-it-most/

2026-01-18 13:18:27 CET

This is exciting. Bitchat - MeshCore bridge. Using Bitchat to chat with people across the city. But this blew my mind. My bitchat message was heard through the bridge on the other side of the country.
400km distance.
Now we need to test it in some crazy dictatorship that turns off comms and the revolution uses it to communicate. Who is smuggling lora devices to Iran and distributing the Bitchat APK?

2026-01-17 22:42:31 CET

Goodbye Firefox. You had a good run. You have now become completely unusable on both phone and desktop.
I wanted to blame the Rust rewrite. It's probably not it, but I can pretend it is.

2026-01-15 11:49:02 CET

Migrating to a new phone is total pain.
Give me back the one day of my life.

2026-01-12 09:26:56 CET

If your time is not valuable, use your time to make your time valuable.
Learn a skill, make a kid, whatever. Time is the only truly valuable and scarce resource.

2026-01-02 21:06:58 CET

- reply

The promise of Nostr is we are not jerks cancelling others.
He created something, you are just yelling at the cloud. Mute the bot if you don't like it and carry on.

2025-12-21 14:36:08 CET

- reply

This is perfect for Christmas shopping in Lidl! Thank you.
https://fountain.fm/episode/l5C7qTbiDlwufRxpegLY
#nevent1q…rjf3

2025-12-08 13:38:16 CET

If the official X app stopped working for you (f.e. if you are using @nprofile…duwt ), or your banking apps aren't working anymore, the problem might be hardware-based attestation. Unfortunately, despite winning many battles for an open internet, open code, open protocols, and file formats, hardware attestation is a problem that most of us aren't even aware of.
And it's a big problem for the Internet.
https://juraj.bednar.io/en/blog-en/2025/12/08/the-internet-and-computers-arent-what-they-used-to-be-why-your-apps-stopped-working-hardware-attestation/

2025-11-02 17:12:59 CET

- reply

Yes, but they will have an earpiece with a synthetic voice telling them how exactly to lay the brick down. "One millimeter to the left, and align. Well done, now the next one,"

2025-09-26 21:25:42 CEST

Now thinking about it, you can probably fully sync and validate the time chain without actually seeing any of the full transactions.
If you have a prover that converts transactions into zk proofs, revealing only unspent outputs, tx hashes (for block hash / pow verification).
You can sync to current utxo set without seeing any of the full txs and their scripts and without trusting anyone, you just follow longest pow proofchain.
I don't think it's very useful (harder to verify zk proofs than the timechain itself). It does not add privacy because the prover sees all the txs. But it's mind-blowing that it can probably be done.

2025-09-05 17:53:05 CEST

- reply

Yes, you can do swap, with the problems I quoted above.
It is as private because of two things:
- you can have as much privacy as you want and you can want same privacy as Monero. No public record of transaction anywhere. And that is regardless of sending and receiving counterparty's choices
- you can actually use it in many more cases, making your financial life more private. I remember at one point a Monero meetup where people were sitting at the bar without drinks. I came to ask if they're ok and they reluctantly said they can't buy drinks because the bar only accepts Bitcoin. So I got them drinks. Many Monero maxis would not do a swap (not very private even, the provider sees the details of the tx, even though they can't identify either party), and especially online, the choice is Monero or credit card. If Monero is not accepted, they pay with credit card. My choice of practical private online transaction is just bigger. I'm not a Bitcoin maxi, I'd gladly pay in Monero and I actually like to do it, but I simply don't have the chance that much. The shop I use most for online purchases (Alza) takes lightning, shitcoins like usdt, but no xmr. I can do swap, but I pay additional fees, I have to wait and hope the swap goes through before the invoice expires (not lightning invoice, just the general payment code expiration due to exchange rate). So for my practical purchases, Lightning is just so widely accepted that it does not make any sense for me to look for other merchants that accept xmr. There's no sender information in lightning transaction, so I just pay and it's private and instant.
So practically, lightning is what gives me financial privacy. Monero is just too niche to work in practice.
BTW: for swaps, I started using Zcash, thanks for your howto, it works pretty well.
I've also done the anypay.today project for this. You scan a QR code, it gets converted through trocador, the source is always lightning because I don't want to wait for miners in order for the payment to go through. One click, instant confirmation and whatever shitcoin the recipient wants they get. With Monero swaps, it's often 10 or more minutes until the recipient sees the tx.

2025-08-29 13:08:38 CEST

Bitcoin on chain: collateral, long term storage (hodl), backing
Lightning: plumbing and pipes, interconnectivity, payment protocol
Cashu, fedi, liquid, lightning, ...: peer to peer electronic cash.
Oh, but the whitepaper said Bitcoin is peer to peer electronic cash
Oh, but you said lightning is going to be used by the end users, not as interconnectivity layer.
Yes, software, technology, narratives evolve. And it's better than if the whitepaper was right and frozen in time.

2025-08-11 10:41:00 CEST

When I had a company managing server farms, I wrote out own ssh client that was brutally scriptable. You could say login here and it would know that in order to get there, it needs to login somewhere else, forward ports, then write su and type the password because of some company policy of the client not allowing direct root logins. We had scripts checking everything and just connecting wherever it was needed without a care in the world.
Then there was version two written by someone else that also logged the content of all sessions to a server so we could show the customer what we did or did not do.
On the plus side, now I login mostly to my machines and I can set it up however I want.

2025-08-10 11:24:34 CEST

- reply

What is it?

2025-08-09 22:47:46 CEST

- reply

tmux can do probably much more than screen. you can even code plugins.
vibe code all the things

2025-08-09 17:04:40 CEST

- reply

Omgomg
I thought I wanted a new terminal app, but I really wanted a tmux plug-in. Since I'm in the process of tuning tmux.
https://tmuxai.dev/getting-started

2025-08-03 14:02:44 CEST

I have started more than 15 companies in my lifetime. A few of them successful. I would do an AMA, but I'll answer three questions that you didn't dare to ask instead.
Q: What is the best advice you have for young entrepreneurs?
A: Stay away from EU. Don't incorporate there. Don't target EU customers.
Q: What is the best target market that no one sees?
A: The netizens that own cryptocurrency and can use them for payments (no ETF owning grandmas).
Q: Where do I find investors or money?
A: Don't. Bootstrap.

2025-05-28 10:00:59 CEST

Ok. So Square merchants can accept Bitcoin payments by default. They get paid fiat if they don't want to deal with Bitcoin (default on) or keep the Bitcoin (if they opt in). This adds four million Bitcoin accepting merchants.
When will this happen in European union? Never, because of MiCA. But EU has an innovation compass!

2025-05-25 11:38:16 CEST

- reply

Ufff, light depka od Hundrátora.
#nevent1q…clu0

2025-05-09 08:50:59 CEST

- reply

Nooo, they will not ban cash, you conspiracy theorist! See, you can still buy condoms and wine anonymously!
#nevent1q…cu8x

2025-04-03 09:02:54 CEST

- reply

That turns web into 1990s style BBS.
Very cool for about two days.

2025-02-24 11:31:45 CET

- reply

Ja sa bojím že je to determinované zvykom. Teda je dosť možné, že o pár rokov si na to zvykne aj väčšia Európanov.
Rád by som sa mýlil.

2025-02-16 22:10:10 CET

- reply

curtains, chairs, bed, towels, trashcan, light setup, ...............

2024-10-27 13:56:04 CET

SK: Predikčné trhy predstavujú praktický spôsob, ako môžeme nahliadnuť do budúcnosti bez nutnosti spoliehať sa na tradičné, často nepresné metódy, ako je veštenie z kávových usadenín, hlasovanie, či komentáre “odborníkov” v novinách. Filip poukazuje na fakt, že tradičné metódy získavania informácií, ako sú prieskumy verejnej mienky, môžu byť dosť zásadne nepresné. To znamená, že predikčné trhy, ktoré umožňujú účastníkom investovať do svojich názorov, poskytujú presnejší a realistickejší odraz skutočnosti
Prečo “gambling” hovorí niečo o svete, ako ho možno používať na hedging budúcnosti, poistky, získavanie informácií, či robenie lepších rozhodnutí alebo dokonca financovanie verejných statkov?
https://juraj.bednar.io/podcast/2024/10/27/filip-legeny-predikcne-trhy-ako-nevestit-buducnost-z-kavovych-usadenin/

2024-10-08 14:51:59 CEST

Let's talk about value4value. I listen to and get value from many podcasts. I'm ruthless and stop listening if they are boring. So when I listen, they are valuable to me. And it does not always mean mind-blowing tech stuff, it can be people who get high and play music.
We have this space age technology where we can stream money, directly to producers of content we value.
And then we send them two cents. I have produced more than a hundred episodes. Many of them have changed lives, literally. My total earn from all value4value was around 35$ - streamed sats and boosts. I don't have sponsors, even though I sometimes promote my own projects. Just the cost to produce one of these is 200$ or more (at least in my time, sometimes I rent a room / studio, although I'm pretty DIY). The cameras and microphones are not free either. Neither is server hosting (I self host, no cloud services, I run my own node).
I am not complaining though. I'm part of the problem. I did not change the default in the app - 10 sats per minute is ridiculous.
I changed my setting to 200 sats per minute to start with. 7 dollars per hour makes sense to me. It's cheaper than the tip I leave for a dinner. But if you have a few listeners that set it up, it might even pay for the production cost.
Your situation might be different. But let's use this space age technology for real, not as a proof of concept to send pennies.
Thanks to @nprofile…0fdq for making me see this clearly.
Repost if you do the same. You don't have to do 200 sats/min. Do whatever feels right. It's probably not the default.
#v4v #value4value #podcasting2.0

2024-10-02 21:44:48 CEST

Congrats to 25 years of Euro!
https://image.nostr.build/ab604ca16f31ea24518205333863798d64312193c69fee0d4f9a3067fd4fe045.jpg

2024-08-18 23:18:34 CEST

Venice.AI is a service that operates open-source image and text generation models in the cloud for end users. They offer a reasonably priced subscription service for state of the art image and text generation models (similar to ChatGPT, Midjourney, Claude). But there is a (legal and supported) way to get a lifetime Pro account without a subscription (do you hate subscriptions as much as I do?), for one-time cost (by buying a MOR token, which I'll simplify for you by just selling you the account where all this is activated).
Check out Venice (it's really cool) and if you want a Pro account, I have them for sale!
Get yours here:
https://pay.cypherpunk.today/apps/26zEBNn6FGAkzvVVuDMz3SXrKJLU/crowdfund
https://cypherpunk.today/static/venice-pro.jpg

2024-04-05 22:13:12 CEST

- reply

yes, but at the same time the same kind of people promote sending weapons to Ukraine.
that contradiction is what is interesting to me

2023-12-27 21:15:25 CET

- reply

Bitcoin not discussed. We run a reticulum network on Lora with a few people, so that's fun.
Otherwise traditional German hacker leftie vibe 😀

2023-08-29 21:13:51 CEST

One of the best life hacks is adopt the following rule in life:
Sauna is always a good idea
/
Sauna je vždy dobrý nápad