241–250 of 913 posts
[flagged]
You don't think Steve Jobs was proud of the iPhone, despite others doing most of the actual physical work on it? I'm more than a mere typist; there's a skill to getting a usable, useful app out of an AI.
There is no skill to using LLMs, you can learn to be effective with them just by using them for a few days.
(Edit: forgotten in first edition) A cookbook to store the recipes my family likes to cook so I can eventually break up with Pinterest: https://github.com/vtbassmatt/Cookbook
A data extraction pipeline and search engine for a new card game called Mood Swings: https://moodswingsdata.github.io and https://moodswingsdata.github.io/feelings.
An app to let my friends and me build a Magic: the Gathering cube iteratively together: https://github.com/vtbassmatt/popcorn-cube
A custom wiki engine for a family of podcasts I enjoy: https://github.com/vtbassmatt/totalus-wikium
A systemd log viewer for the web: https://github.com/vtbassmatt/djournal
Created a few Open Source tools: Open Camera Control https://github.com/jcubic/open-camera-control - that allows me to control the settings of my DSLR while I'm recording myself. Horavox - https://github.com/jcubic/horavox - A speaking clock Mutimon - https://github.com/jcubic/mutimon - a config driven web scraper (found this post from the email sent by this tool). ASCII-Globe - https://github.com/jcubic/ascii-globe…
I need to finish off that blog post. With Tailscale, you can basically point a domain to the FQDN of a machine you’re sharing with people and the domain will simply work for them (and only them)[1]. But for it to work without them having to know or specify the specific application port, you have to grant them access to 443 (and 80) in your Tailscale ACL for that specific host. So yeah, now immich.familydomain.com wor…
Eg: push a button, it shows that it's working for a while, then strongly flashes when it's done (success/failure). When you have it right under the monitor, it's like a macro pad for long-running things.
This reminds me of some of the very early peripherals you'd see on the Alto and other computers. I was surprised something like this didn't seem to exist, but maybe I'm just terrible at searching.
I need to finish off that blog post. With Tailscale, you can basically point a domain to the FQDN of a machine you’re sharing with people and the domain will simply work for them (and only them)[1]. But for it to work without them having to know or specify the specific application port, you have to grant them access to 443 (and 80) in your Tailscale ACL for that specific host. So yeah, now immich.familydomain.com wor…
authentik is also an oidc provider... couldn't it also be the way they auth to tailscale too?
If tomorrow I decide not to share with anyone, I don’t want to have to reconfigure stuff. I simply edit the Tailscale policy file, and (maybe) spin down my server.
* Codjiflo: A code review tool inspired by Microsoft's CodeFlow: https://codjiflo.net
* A virtual replica of a digital readout (DRO) for operating a CNC machine like a manual mill: https://el400.vza.net
* Reverse engineered CNC pendant integration with CNCjs also for operating a CNC machine like a manual mill: https://github.com/pedropaulovc/cncjs-pendant-whb04b-6
* A 'docker compose' to provision email, chat and documents for human-AI hybrid teams where you can take over AI's agent's credentials temporarily: https://github.com/vezzadev/roster
* The CNC stuff will come handy for a bigger project I have to create a 1:1 replica of Albert Michelson's harmonic analyzer: https://github.com/pedropaulovc/harmonic-analyzer
* Reverse engineered Hik-Connect P2P CCTV protocol for integration with OSS like Home Assistant and Frida: https://github.com/AlexxIT/go2rtc/issues/2289
* Some patches for different OSS projects like improvements to MCP tools, Playwright, Claude Code, etc.
An attempt at an artificial unconscious. Turns out it’s pretty hard to inspire an LLM to be creative. https://sisuonspeaks.com/
Do you have a write-up on how it works under the hood or is it meant to be more mysterious?
I wasn’t planning on posting it yet, so I’ll have to get you a thorough follow up, but in a nutshell it’s: a continuously running 448-concept space (philosophy, cognition, art, nature, math) that occasionally “crystallizes” a group of 4 concepts via Hebbian learning and stochastic noise. Those concepts get sent to the LLM with minimal guidance beyond some safety guardrails and encouragement to be creative. It has access to a sandbox to produce essays, stories, music, “art”, and small python scripts. Self-updating memory system. Notes, essays, and artifacts can also be discussed with me through some outbox channels.
On top of that it’s got a separate academic philosopher + psychiatrist llm that critiques its work and has a regular cadence of “sessions” with it, as well as a research assistant bot who I talk to (but doesn’t interact with Sisuon) who has full project context and memory access. The sisuonspeaks site is a VERY abridged collection of Sisuon’s essays, along with analysis, commentary, forum posts, and a podcast…all created by, you guessed it, more LLMs.