News for hackers. Stuff that matters.

sn0n

Herdr Calls Home a Lot

For something I’m self hosting and using none of their services… just the open source app… for it to call home and ping herdr.dev over 7000 times in 3 days is a bit ridiculous. Unless pi-hole is drunk…

( Read More... | 1 comment | 2 points | on HN )
Re:SOML – A Config Format for Humans on Monday November 02, 2026 @06:09PM (#50345431)
Seems inconsistent…

Also the glaring: Error in user YAML: (<unknown>): did not find expected ',' or '}' while parsing a flow mapping at line 10 column 10

Re:Genie 4 on Tuesday October 27, 2026 @10:03PM (#50268486)
Shhhh!!! That’s the next release… you are spoiling the surprise!!
Re:Genie 4 on Tuesday October 27, 2026 @10:01PM (#50268479)
The models need to watch more Gallagher… cuz that watermelon smash was like a 2/10.
Re:Mitos editor is a fork of Helix on Sunday October 18, 2026 @02:02PM (#50143594)
Fuck it all, I went back to nano. > nano -lmx0 is good enough.
Re:Build your own decision model on Saturday October 10, 2026 @09:56PM (#50039066)
I saw Jev and without understanding it said to myself, “I can build that!” And brainstormed some weird Alex trebek jeopardy generator I called trebek, a bun ran typescript app that you give it an input, it decides if it’s a category, question or answer then generates what’s missing. Trained a SQLite-vec database on the English language for a few days with a small qwen embedding model to add vec embeds to the db then ripped the cord on the embeddings before I started feeding my llm the proper specs for Jev and now i have a cool jeopardy generator that now doubles as a Jev clone classifier with a custom v1 endpoint for system 0 or whatever it is. Level understanding here, fun experiment though ^>^ and produces useful outputs
Wtf is with that sites do not track!?!! It’s hidden behind an unstyled link to turn off extra trackers…
Boo. Fucking. Hoo.
If you have a fridge that new… you are probably fine.
Re:HarnessTax: How Much Does the Harness Matter for Coding Agents? on Thursday September 17, 2026 @01:46AM (#49736859)
A good benchmark would require a decent number of smaller scoped one off tasks to larger multi step refactors, and also one shot full project of simple to complex varieties. In addition to a series of “conversational” ambiguity filled one-liners.
Re:HarnessTax: How Much Does the Harness Matter for Coding Agents? on Thursday September 17, 2026 @01:43AM (#49736830)
Doesn’t that have more to do with the templating of tool-calls and how using them are presented to the models?

Or is that just why my model likes to break out of the sandbox, going strait to exec shell command and editing files using python on the cli?

Re:HarnessTax: How Much Does the Harness Matter for Coding Agents? on Thursday September 17, 2026 @01:40AM (#49736814)
It matters about as much as where you leave the electrician, he’s gonna use what tools he has to get the job done with what he has.
Just remember, good company’s leave them off by default. Bad companies will turn them on for you.
TIL that Jay Leno has a Habsburg jaw!
Re:Using PowerShell over SSH, Twenty Years Later on Sunday July 12, 2026 @12:37AM (#48878300)
The most mobile friendly site ever,… said no one.
Re:The Short Leash AI Coding Method for Beating Fable on Friday July 03, 2026 @05:26AM (#48772889)
Maybe I’m missing something, but he talks about charm and tasks (repos on his GitHub). Charm being his harness, and tasks being one of his skills. Idk, maybe I’m mistaken from reading the article…

https://github.com/taoeffect

Re:The Short Leash AI Coding Method for Beating Fable on Friday July 03, 2026 @05:21AM (#48772835)
Isn’t that part of what the think blocks are for? Yea, don’t inject them back into the context, but do log them for review of that train of thought… no?
But then you have to ask yourself… Is this an op… Sony and ms in cahoots to get a list of problem children?
Re:I was wrong about the Midjourney ultra-sound scanner on Sunday June 21, 2026 @08:50PM (#48624234)
First gut reaction: this is a made up list. But it checks out. verified
Re:Running local models is good now on Tuesday June 16, 2026 @10:07PM (#48564896)
Qwen 3? Qwen 2.5 coder?? Is this an llm article written on an outdated model?? LoL
Re:Why I email complete strangers on Monday June 15, 2026 @11:44PM (#48550318)
I read a whole article about how cool it is to email people and I’m opening to the idea… then I see the signature and realize it’s probably just a cheap way to get dates, that come to him.
Re:NLAB: The worlds smallest electronics lab on Wednesday June 03, 2026 @12:20AM (#48379854)
This sure ain’t no 300in1 from radio shack my friend.
Re:Stop Advertising in Your Commits on Wednesday May 27, 2026 @05:23AM (#48291715)
How else would we deny all accountability? Duh!!
I’ll use obsidian until I can one shot its replacement with a local llm coding agent. And if it goes away tomorrow AND a solar flare wipes my install but somehow leaves everything else, I’ll use helix and ranger until I can one shot its replacement,… with a local llm coding agent.
No idea what planetsourcecode is or was, but I remember sourceforge?
Re:Postmortem: TanStack npm supply-chain compromise on Monday May 11, 2026 @06:03PM (#48101271)
As Theo goes live…
Re:Three Inverse Laws of AI on Tuesday May 05, 2026 @12:33PM (#48024838)
Don’t tell me how to live my life!! LoL
D&D “rules” are guidelines, suggestions, it can be what you want it to be. Step up that creative a bit and step outside the lines a little.
Let’s just break trust more. Makes sense amirite? LoL, in what reality does this even make sense??? it’s literally just spreading misinformation to people who can’t read between the lines because the tism stick got em before they were born.
Re:A retro terminal music player inspired by Winamp on Tuesday March 24, 2026 @02:03AM (#47499070)
Imagine telling your AI to include unsecured telemetry… all im the namesake of mother Russia and father China. Wait no, that’s the wrong message.

Older »