This is covered in Claudeish, had to stop reading. I have to read enough of this at work...

>What sits behind them is a calibre I had to solve the way a real one is solved, and this is why I bother.

>A skeleton is a load-bearing machine with the metal taken away down to the last safe gram, so what is left is whatever carries a force, and no spoke is wider than its job.

>Running and buildable are different questions, and the second is harder.

>A drawn movement is a claim. Before anything gets a surface, I test it as a machine.

If this isn't Claude-prose then the author has been reading too much of it...

(oh, Pangram gives the entire thing 100%)

I think "load-bearing" and "surface" is what gives it away. It continues to perplex me how something trained on human-written English ends up producing such output. It's much worse than the classic em-dashes and "it's not X, it's Y" (but maybe that was a different LLM?)

I didn't recognize Claude but the thing was unreadable and after a while I figured out it must be AI.

Give them some grace. The author is a Czech developer likely using AI to help write in a second language or translating.

If the whole thing is written by AI, then I wouldn't say that it was just "helping". Author could write it in their native language and then use AI to translate it - the result would be very different.

The movement does look pretty neat. I'm not into watches, but it's impressive that purely rendered thing was designed to be realistic.

Google Translate usually writes better than this (closer to ESL-grade English, IMHO).

Interesting, but title is missing the key word "Garmin".

"How I build a mechanical Garmin watch face: a real gear train for a watch with no gears"