Tootfinder

RE: https://hachyderm.io/@numb_comfortably/115924798786314393
The thing about this is: “AI” can neither review anything nor can it make mistakes.

I made the mistake of asking Claude to suggest some updates to my steganography lecture and now I am stuck reading an inch of papers this weekend.

Just made a mistake on page 40 of a #LEGO build that turned out to be wrong, but it didn't interfere with anything until page 242. It was a but janky, but thankfully I didn't need to unbuild everything in between to fix it. Lego puts a huge amount of attention into the design of their instructions, but things like this are pretty hard to catch. I was building quickly and assumed a beam should go on symmetrically when actually it needed to be off-center.

I'm sure there's more to switching party than just taking off your stupid false moustache.

I’ve been working on some Python code for a few days now… just an hour or so in the evening as time allows. I could probably use some AI Slop Generator to quickly kick out what I need, but that’s not gonna happen.
No, Nope, No Way.
I actually *want* to write the code, and understand it, and know that every choice was mine, even if I make mistakes along the way.
If I publish it I want it to be from me, not from me and the AI Slop Generator and every codebase it ever stole …

The Demon-Haunted World: Science as a Candle in the Dark by Carl Sagan is a non-fiction book published in 1995.
I just finished reading it and I swear to you it is as relevant today as when it was written. As if 30 years haven't passed. Humanity keeps repeating the same mistakes over and over again and again.

I would like to humbly recommend that every SOC have a giant sign saying “Absence of evidence is not evidence of absence unless you’ve verified the logs would be there” This is such an easy mistake to make and it leads cascades of communication failures

TIL that WB0RLJ Jim posts videos of his POTA activations keyed by time, so you can find your call in his video description with a link straight to the time in the video. Very cool!
I also read his QRZ page and discovered that I did something that he dislikes. I've heard other ops say that if someone gives a partial call, you should reply with your sign twice to cut down on back-and-forth, and now I kn…

AI can’t make mistakes*, but you can make the mistake of believing AI can make mistakes.
*they don’t have a mind, cannot think, don’t conceptualize and have no understanding of anything, therefore they can’t be mistaken