ynniv

i've spent decades reading and writing code. i just want something that works

I have a work-in-progress called Modus. 100% written by Claude, so take that however you will. The current release boots on a Raspberry Pi Zero 2 W. The next release (unreleased in the pipe for ~ months) is standard Common Lisp on bare aarch64 (pi) and x64 (qemu for now), with linux aarch64 and x64 command line interfaces à la sbcl.

https://github.com/modus-lisp/modus

Since you can't use an OS by itself, I've rounded out the Common Lisp environment with portable ssh client and server, web browser, and a bitcoin node. Framebuffer with VNC in the pipe

+1 Having a travel time speed of [∞, realtime] would make this installation art

damn, that's an unintentionally amazing advertisement for comaps. downloading now! X-D

nostr

it would seem so. so the question is why they would maintain an air gap for a safety report

pretty weird...

  NATIONAL TRANSPORTATION SAFEN'BOAFID
  :J11...:i= of ArutiOn SMel@LA5 301
  A.,r Tral1C.IYU';UQlt-Ort!!
  NTS,B ri@ss @at.Joo JurtJer
  DCA26 22\C2<

eclipse also does better than dot, but it's useful to notice that graphviz never dies

https://github.com/eclipse-elk/elk

personally i find llms to be absurdly capable, and people are just using them wrong

[flagged]

i'm in atlanta. i used google. it took five seconds

try google. it took 5 seconds

a leaked document shows that at least one person was recognized to have the appropriate visa and was coerced into leaving "voluntarily". no details on how prevalent this was have been released

https://www.theguardian.com/us-news/2025/sep/10/hyundai-fact...

surrounded by spaces... right?

Finally some hacker news

https://github.com/block/goose

  Dog
  Cat
  Jesus

point eight bee pee ess, said no one ever. Also too much ADR to censor words that didn't need to be censored

"There isn't a government on this planet that wouldn't kill us all for that thing"

Once you accept that a movie isn't a documentary, Hackers is a lot of fun. The editor of 2600 consulted on it (and lent his nym)

Who sees an original mouse and thinks "speech recognition" without this classic context?

Only formally proven systems will be secure

You could try issuing two prompts: one to "think", and another to "say to the group". Not as nice as having a model that can properly format its response, but you'll get more accurate results.

This is a fantastic idea! Reading through one of the games I see:

- players are named based on their model, which can be ambiguous

- some model responses are being cut short

- some models seem to be thinking out loud, or at least not separating their chain of thought from what they tell the group

It better be! AMD @ $2k: https://digitalspaceport.com/how-to-run-deepseek-r1-671b-ful...

This is excellent

For a second I thought it had been ported to wasm! This is still a browser based VNC

Everyone should be interested in Qubes OS.

Is the clipboard also a "data leak"?

"you may have concerns about privacy"

Way to FUD: these models are local unless you enable ChatGPT integration.