<unnamed>

Public Key

npub1jlrs53pkdfjnts29kveljul2sm0actt6n8dxrrzqcersttvcuv3qdjynqn

Profile Code

nprofile1qqsf03c2gsmx5ef4c9zmxvlew04gdh7u94afnknp33qvv3c94kvwxgspp4mhxue69uhkummn9ekx7mqpr3mhxue69uhksmmyd33x7epwvdhhyctrd3jjuum0vd5kzmqqdlqcs

Publishing to

nos.lol

hodlbod.coracle.social

hbr.coracle.social

relay.damus.io

nostr.wine

Last Notes

2026-03-08 16:19:18 CET

- reply

Awesome. My sone came and sat next to me and made this:
https://hbb.coracle.social/5564c00d280fe0209c11776814886c23b39e4bce44c0b7ff603053b4ab17f708.jpg

2026-03-08 14:41:39 CET

I made a spoon
https://hbb.coracle.social/7b63558d4e2a008cd1029f9806a3d2b878c0551112a575d2b4a798ea6ccf3a1b.jpg

2026-03-07 01:25:39 CET

@nprofile…jejf I gotta ask, how is wisp so fast

2026-03-07 01:13:29 CET

switching to @nprofile…6nzx makes spending on tokens hit different, I've spent 300k sats this week 😅

2026-03-07 01:08:05 CET

@nprofile…jejf I love the custom colors, and the post delay interface, great work on wisp

2026-03-07 01:07:16 CET

- reply

looks good, but I think it loaded after I quoted it, one of those things

2026-03-06 23:57:34 CET

- reply

There, this is better
https://hbb.coracle.social/6b656516440227dc1d470447907f6db239da89fcf1911bd27d3d831d64151052.png

2026-03-06 23:47:34 CET

- reply

I considered that, but it's very bulky and you may have more than 3 depending on how signers are set up. But I have an idea for improving this

2026-03-06 23:34:52 CET

Getting close to being able to release pomade to Flotilla users. Here's a boring video which showcases the UX:
https://coracle-media.us-southeast-1.linodeobjects.com/pomade_demo_20260306.webm

2026-03-06 23:07:40 CET

- reply

Nice! I will probably use this, it could also work great for moving stuff to untrusted devices using one-off sign in with nip 46

2026-03-06 21:52:58 CET

- reply

Would be funny if the withholding of copyright was the thing that made everyone go full open source

2026-03-06 17:18:32 CET

- reply

Hints all look ok, here's the note:
#nevent1q…cvk8

2026-03-06 17:12:09 CET

- reply

It never did load, but it looks like it was only sent to mleku's inbox relays for some reason. Could be a coracle bug

2026-03-06 01:26:34 CET

- reply

This is very good

2026-03-06 01:05:55 CET

- reply

Oh, I see, yes I ignore <15

2026-03-06 01:05:13 CET

- reply

I see a 16 on this note

2026-03-06 01:04:47 CET

My spirit animal is this website
https://406.fail/

2026-03-06 00:10:35 CET

- reply

Debian's days are numbered

2026-03-05 23:14:17 CET

- reply

I don't know how true this is, but maybe because they see it as competing for funding with bitcoin, as a distraction, or even as "crypto"? I know there was some talk about this back when OpenSats started their nostr fund.

2026-03-05 23:11:50 CET

I've heard people talk about the technique for agentic coding where you have a bunch of agents build the same thing in parallel 100x then pick the best implementation. I'm sort of doing the same thing with pomade right now — instead of building one implementation of the protocol, I'm building three (one in typescript, one in rust, and one in go). But what's neat is I don't have to choose one, because the whole idea is to have multiple separate unrelated custodians, each person can run an entirely separate codebase.

2026-03-05 21:59:21 CET

- reply

Me too. I've used it to good effect to write some code in a domain I don't know in languages I don't know, and I can mostly verify the results by writing tests. But for domains and languages that I do know, I much prefer writing code manually (and I think it's better and faster).

2026-03-05 21:58:06 CET

- reply

it's beautiful

2026-03-05 21:28:53 CET

> It’s been six years since GPT-3, and there has been no “move 37” moment for writing (as there was for AlphaGo’s creative play of Go). Not even close.
https://www.theintrinsicperspective.com/p/bits-in-bits-out

2026-03-05 19:36:45 CET

This article says a lot of what I wanted to say about LLMs, but couldn't find the words: https://acko.net/blog/the-l-in-llm-stands-for-lying/
I don't agree with his conclusion that leaning in to intellectual property rights and source citation is the solution, though that's an interesting though. But there are some great sections, particularly in the first half. Here are some highlights:
> LLMs do something very specific: they allow individuals to make forgeries of their own potential output, or that of someone else, faster than they could make it themselves.
> Experienced veterans who turn to AI are said to supposedly fare better, producing 10x or even 100x the lines of code from before. When I hear this, I wonder what sort of senior software engineer still doesn't understand that every line of code they run and depend on is a liability.
>
> One of the most remarkable things I've heard someone say was that AI coding is a great application of the technology because everything an agent needs to know is explained in the codebase. This is catastrophically wrong and absurd, because if it were true, there would be no actual coding work to do.
>
> It's also a huge tell. The salient difference here is whether an engineer has mostly spent their career solving problems created by other software, or solving problems people already had before there was any software at all. Only the latter will teach you to think about the constraints a problem actually has, and the needs of the users who solve it, which are always far messier than a novice would think.

2026-03-05 17:25:19 CET

- reply

Sounds great! Another feature request is to add api request history somewhere. On opencode I spent $20 very quickly and it was really helpful for figuring out exactly which API calls caused it. Including number of tokens in/out, dollar amount of upstream, and payment amount would be awesome. Also, I've noticed that the current tally of dollar amount doesn't show in opencode like it does for other providers.

2026-03-05 17:21:41 CET

- reply

Nice, sounds like a great way to simplify. I like routstr a lot, but here are a few other things you could to do improve the UX:
- Don't process payments synchronously in chat
- Auto-refresh balances in the background?
- Maybe eliminate the balance on a key? Although having a budget on a key is nice.

2026-03-05 03:16:34 CET

- reply

I may be wrong, I tried again with another pubkey and it didn't merge accounts

2026-03-04 23:10:41 CET

- reply

delusional

2026-03-04 23:08:44 CET

Just added this to my opencode build prompt:
> You are in a docker sandbox, which means timestamps on files are often incorrect. To get around this, always touch a file before editing it.
🙄

2026-03-04 21:53:44 CET

- reply

Yeah, pretty nice. Although it auto linked my email account, could there be an attack vector based on non-unique names?

2026-03-04 20:52:18 CET

- reply

Got it working, thanks!

2026-03-04 17:54:06 CET

@nprofile…uy5c https://git.mleku.dev/mleku/gitea-nostr-auth is giving me an SSL error

2026-03-04 17:50:56 CET

- reply

Coracle isn't really the place to showcase dvms, custom feeds notwithstanding. Maybe something like noogle.lol would be better

2026-03-04 01:55:15 CET

agents just introduce a whole new category of yak shaving

2026-03-04 01:55:03 CET

Me: follows the directions for setting up NanoClaw on a fresh VPS
NanoClaw:
```
lsof /var/lib/dpkg/lock-frontend 2>&1
kill -9 19631 19750 2>&1; sleep 1; rm -f /var/lib/dpkg/lock-frontend /var/lib/dpkg/lock /var/cache/apt/archives/lock 2>/dev/null; dpkg --configure -a 2>&1
```

2026-03-04 01:52:19 CET

- reply

How is sprites different from just setting up a VPS? How's the latency?

2026-03-04 01:51:13 CET

- reply

I have been thinking about writing a client that would timestamp messages as sent on the current date in 2021 or something. could be fun, partitioning content by epoch. It would also screw up any idea of the authenticity of old posts

2026-03-04 01:46:34 CET

- reply

You clearly never wrote php in 2012

2026-03-04 00:47:58 CET

Vibe coding is the death of abstraction. Why use the visitor pattern or transducers when the LLM will just scatter `if` statements everywhere anyway

2026-03-03 20:44:57 CET

- reply

That 20% matters. I try to use kimi k2.5 as much as possible, but it's just not as good as sonnet

2026-03-03 20:44:11 CET

- reply

stuff != wealth

2026-03-03 19:42:50 CET

- reply

I don't know, I don't really use this stuff. I can just smell the spaghetti from around the block

2026-03-03 19:42:04 CET

Agents are getting more expensive:
https://newsletter.danielpaleka.com/p/you-are-going-to-get-priced-out-of

2026-03-03 19:21:51 CET

- reply

No one should use openclaw, use nanoclaw or ironclaw or something else if you must run an agent

2026-03-03 18:42:52 CET

- reply

#nevent1q…sa30

2026-03-03 18:42:37 CET

```
docker sandbox save opencode-flotilla my-opencode:v1.0.1
Snapshotting image in sandbox ...
Reading image from sandbox ...
Save complete. To use the image:
docker sandbox create --load-local-template -t my-opencode:v1.0.1 [...]
```
```
docker sandbox create --load-local-template -t my-opencode:v1.0.1
unknown flag: --load-local-template
Usage: docker sandbox create [OPTIONS] AGENT WORKSPACE
Run 'docker sandbox create --help' for more information
```
great, thank you docker
#nevent1q…jl3k

2026-03-03 18:41:02 CET

docker sandbox is clearly vibecoded. Half the flags don't work, and half the commands advertise flags that don't exist.

2026-03-03 17:38:31 CET

- reply

what is UTM?

2026-03-03 03:51:14 CET

Spent the day fiddling with agent isolation. At first I went down the rabbit hole of setting up a dev environment on an old macbook and acc,ssing it over wireguard, but the latency was annoying. Then I tried matchlock, which was promising but had weird build and control character issues. Finally, I went with docker sandbox, shich is good enough, although I had yo use a very dumb hack to get my config into the container. This is a massive product opportunity.

2026-02-28 02:41:51 CET

- reply

They're not autonomous, I write the prompts

2026-02-28 02:07:56 CET

- reply

Just coding

2026-02-27 23:12:32 CET

- reply

nooo no no haha just opencode

2026-02-27 23:07:53 CET

- reply

Setting up nix on an old 2012 macbook I have lying around. We'll see if it survives 😂

2026-02-27 22:30:55 CET

- reply

Yeah, I was thinking about that. I intend to give it all its own keys.

2026-02-27 22:30:46 CET

- reply

Do you ssh in and have all your code on the box, or send context from your machine to it somehow?

2026-02-27 22:08:50 CET

- reply

I bought some old laptops from a school on ebay, 5 for $100. One of them doesn't work, but it was a great deal.

2026-02-27 22:08:21 CET

Now that I'm using agents more extensively, I'm thinking about moving my development environment to a VPS to make sure the agents don't send any important data (like my ssh keys) to my provider. Am I being paranoid, or has anyone else done this?

2026-02-27 21:50:44 CET

- reply

yes

2026-02-27 19:59:04 CET

- reply

Very good articulation I hadn't hear before. We're already seeing a lot of those fragility points with the increasing ability to impersonate/scam people in ways that we have zero defenses against. Many more will come I'm sure.

2026-02-27 18:34:11 CET

- reply

He would believe in at least one notable exception

2026-02-27 18:33:35 CET

Coracle 0.6.31 is out:
* Upgrade capacitorjs
* Upgrade nip 55 signer library (fixing Primal Signer)
* Fix crash in UserContent when mute list is empty

2026-02-27 18:19:46 CET

- reply

💯

2026-02-27 18:03:21 CET

- reply

Is AI anti-human, pro-human, or something else?

2026-02-27 17:03:04 CET

- reply

Totally different tech stack, but the only thing I've found that works on web is to load everything from indexeddb on boot and hold it in memory

2026-02-27 00:12:09 CET

- reply

ah no wonder

2026-02-26 22:58:24 CET

- reply

#nevent1q…6l60

2026-02-26 22:57:44 CET

- reply

Thrill seekers are their target market

2026-02-26 22:56:54 CET

- reply

sheesh

2026-02-26 22:38:29 CET

- reply

Other good relays for signers are are ephemeral.snowflare.cc and bucket.coracle.social @nprofile…053e

2026-02-26 22:02:31 CET

- reply

Four years ago, but web of trust is "good enough" so I stopped thinking about it. Targeted attestations are a good idea though

2026-02-26 21:48:21 CET

- reply

Version is 4.1.12, closing and reopening freed it up, but the connection looks halfway broken (Amber shows it but ngit never completes). I do think relays may have something to do with it, ngit is using relay.nsec.app and relay.ditto.pub

2026-02-26 20:48:21 CET

"The heavens, even the heavens, are the Lord's, but the earth He has given to the children of men."
"Thy kingdom come, thy will be done, on earth as it is in heaven."

2026-02-26 20:20:23 CET

- reply

that's not the point, this is:
https://imgs.xkcd.com/comics/standards_2x.png

2026-02-26 20:19:39 CET

- reply

sure

2026-02-26 20:08:50 CET

- reply

Still happens in ngit 2.2.0. I think the problem might be relay selections?
@nprofile…h2ck the loading screen persists on every page even when I navigate.

2026-02-26 20:05:25 CET

- reply

another standard with partial support, neat

2026-02-26 20:04:47 CET

- reply

Easy UX != simple UI, it means things "just work". Do all 100 wallets interoperate seamlessly with the other 100 (10k combinations)? Certainly not. That's my point — not that routstr didn't work (it did), but that it's crufty UI reveals the underlying complexity that leads to poor UX.
We can agree to disagree, I don't expect to convince you. But I do think you need to take this problem seriously in order for cashu to succeed (and I think you do, you just seem to treat people bringing this up dismissively, which is weird).

2026-02-26 19:57:10 CET

- reply

Oh, this is ngit 1.6.0, let me check for an upgrade

2026-02-26 19:55:25 CET

@nprofile…ntqg @nprofile…h2ck I'm getting an infinite loading screen when trying to scan a qr code generated by ngit on both 4.1.7 and 4.1.12. Regular bunker URL generated in Amber works fine (although with a decent delay in ngit, and with the bunker url being printed a bunch of times). Let me know if there's any more info I can provide.

2026-02-26 19:21:45 CET

- reply

And one way to cope is to advocate for dropping support for certain options

2026-02-26 19:20:25 CET

- reply

I have probably 20+, this example was just for illustration

2026-02-26 19:19:43 CET

- reply

These are all obviously much better, but unless all wallets seamlessly support onchain, the various flavors of lightning, and cashu, you're always going to have friction when sending/receiving.
"scan what?"
"receive what?"
"what kind of qr code is this?"
"oh, my phone doesn't support NFC"
"my wallet doesn't do onchain, can you send me a lightning invoice?"
"it's saying something about liquidity"
"what's a 'mint'?"
"what is a 'melt'?"
"what is a 'submarine swap'?"
"why do I need to 'topup', can't I just transfer funds from my bank?"
None of these problems are impossible to solve individually, but solving all of them across a network of apps created by people with different ideas is combinatorially difficult.
In nostr we say "don't overload kinds". I think this is the problem — we've overloaded the term "pay with bitcoin". It means:
- P2PKH
- P2SH
- Native SegWit
- Taproot
- Bolt11 invoice
- LNURL
- Lightning address
- Bolt12 invoice
- Keysend
- Cashu eCash
- Fedimint eCash
- Spark
- Ark
I'm not saying each of these doesn't have its use case, or that innovation isn't good, but this complexity comes at a cost, which either the wallet developer or the user has to bear.

2026-02-26 19:06:08 CET

- reply

I have four wallets:
- Onchain onramp (strike)
- Lightning (alby)
- Cashu balance in routstr
- Token balance for an api key
It's like 2 or 3 too many

2026-02-26 19:04:58 CET

- reply

I try

2026-02-26 18:36:48 CET

- reply

Show me an equivalent UI that is clean and easy for users to understand

2026-02-26 17:52:01 CET

A few months ago I asked @nprofile…xvl9 to sell me on using cashu. His argument was focused on better payment UX, which is compelling.
But something I think is left out of the conversation around lightning/ecash is the funding UX, which can be very complex.
I recently started using routsr (and appreciate it), but it exemplifies this UX friction perfectly. I've written lightning and cashu integrations, and even I am confused by what the UI is telling me. This is not (entirely) the fault of the routstr developers, it's a result of the irreducible complexity of separate protocols for different technical tradeoffs.
https://coracle-media.us-southeast-1.linodeobjects.com/routstr_cashu.webm

2026-02-26 17:31:52 CET

- reply

What else is there to talk about any more

2026-02-26 17:31:20 CET

- reply

Intelligence will still be commoditized

2026-02-26 17:31:09 CET

- reply

Yes, which is why we are different from the bots

2026-02-26 17:30:38 CET

- reply

Which humans then read, as if they're agents. Weird state of affairs

2026-02-26 15:16:52 CET

- reply

Let's talk, I promise I will try not to miss the meeting again 😅

2026-02-26 15:16:21 CET

- reply

But does it have oat milk

2026-02-26 15:13:59 CET

- reply

Yeah, drivers for browsers/emulators/apps would be a game changer

2026-02-26 00:37:23 CET

- reply

😂

2026-02-26 00:37:09 CET

- reply

You're still special, the models can't do it alone

2026-02-26 00:25:00 CET

- reply

You can pull tmux out of my cold dead hands

2026-02-26 00:24:13 CET

- reply

I think most people underestimate the ecological effect of new media. Whatever the journey looks like, new media produce new kinds of people. When that media is powerfully computational, quantitative, artificial, and possible only at scale, it creates people who are a mirror of those attributes. Some may react, some may conform, but you can't go through it unchanged. This could end well, and I expect that because I trust in God's sovereignty. It's easy to imagine many bad outcomes. Only God (or deluded transhumans) can imagine plausible good ones, I think.

2026-02-26 00:16:26 CET

- reply

HEY

2026-02-26 00:16:07 CET

- reply

Sonnet is so good I can't tell the difference. I also don't want to nuke my bank account 😂

2026-02-26 00:15:19 CET

- reply

We are losing
#note1cam…9hdz