Theuni

🏳️‍🌈🏳️‍⚧️

Fortschritt, Nachhaltigkeit, Fürsorge | Progress, Sustainability, Care

🛠️ Aramaki – a federated DevOps control plane

❤️ Cynefin, CAS, Python, NixOS, batou, Ceph, Qemu, Linux, pycountry

... and in a former life: Zope and ZODB

Founder and CEO at flyingcircus.io

github: https://github.com/ctheune
batou: https://batou.readthedocs.io

Last Notes

2024-11-18 09:17:28 UTC

Das Fediverse hat's leider nicht auf die Social-Media-Folie bei der #denog16 geschafft - schade :(

2024-09-20 11:50:18 UTC

Did somebody every do a a redundant array of inexpensive floppies?

2024-09-12 19:06:18 UTC

Looking for a kernel developer who'd be interested to (freelance) help us fix a bug in the mm/xfs area in exchange for €. We're stuck on 5.15 because that little bugger started in 6.1, is apparently still there, regularly produces data loss and is hard to reproduce. #fedihelp #followerpower (Sorry for posting a second time, the first one was in German and I guess that wasn't too much of a smart move.) #linux #kernel #xfs

2024-09-12 18:59:25 UTC

Ich bin auf der Suche nach Kernelentwickler*innen, die sich Freelance im Bereich mm/xfs auskennen und bereit wären für € gemeinsam einen Bug zu jagen der uns daran hindert auf 6.1+ zu upgraden, der Datenverluste erzeugt und hart zu reproduzieren ist ... #fedihelp #followerpower ?

2024-08-12 07:18:28 UTC

- reply

I was joking by impersonating an LLM answer.

2024-08-08 16:01:26 UTC

- reply

Nope, the grifter has no benefit from understanding. It might inhibit him from overselling the item in question. He definitely is better off with whatever half-understanding he has.

2024-08-08 06:40:53 UTC

- reply

I'm assuming you're ironic, but there is a point worth discussion. Mathematical models are intended to cover all general cases. Specifically in computer science we've been bitten by exactly the idea of "nobody will ever put this in" (see y2k bug). We've been through this. And on a more anthropological level good math (and physics) gets extended and remains valuable even in shifting contexts. LLMs are garbage in that way.

2024-08-07 15:01:41 UTC

- reply

It's fantastic. Here's the FastGPT version of it. It greatly demonstrates that it doesn't have the ability to count / compute but only combines textual artefacts.
https://f2.tchncs.de/media_attachments/files/112/921/337/976/559/277/original/3cddeadaebb669c9.png