> An agent has no such learning ability. At least not out of the box. It will continue making the same errors over and over again. Depending on the training data it might also come up with glorious new interpolations of different errors.
This is the main thing that I find annoying about LLMs tbh.
https://mariozechner.at/posts/2026-03-25-thoughts-on-slowing-the-fuck-down/
1. assume every camera has film unless you personally verify otherwise
2. never point a camera at something you're not willing to photograph
3. never put your finger on the shutter button until you're ready to take a photograph
4. always be aware of the subject of your photograph and what is behind the subject
I put all 8,642 Spanish laws in Git – every reform is a commit: https://github.com/EnriqueLop/legalize-es
Discussion: http://news.ycombinator.com/item?id=47553798
22. Now that I have my setup with tge Khadas Edge 2, I kinda want to get a small screen for it for the times that I don't want to wear a screen on my face. Apparently the touch screen works on Linux?
What if ActivityPub software came with a special (@)moderators account that could support instance to instance moderation conversations, as well as a public inbox that anyone could message with reports or related activity.
It would even come with a gorgeous admin dashboard to track reports and conversations with custom labels for organization and webhooks for 3rd party integrations (like getting notified on slack or discord)
🤔
seriously like all the malware google claims to be stopping with the app verification thing is actually on the google play store. i don't think i've ever seen malware on f-droid, but you can simply go to google play and search "flashlight" to find tons and tons of malware
#ProTip Switch your keyboard layout to something like azerty to make your key smashes less predictable
Occult Enby that's making local-first software with peer to peer protocols, mesh networks, and the web.
Yap with me and send me cool links relating to my interests. 👍