Interesting reads
- The hidden purple memorial in your browser - reddit link, the full backstory
-
Dr. TJ Kendon - on what appears to be a new blog - has a wonderful essay on ChatGPT and its best use in schools. Particularly (my emphasis):
By design, ChatGPT and LLMs are knowledge free. They’re designed to take a corpus of text and determine for any particular chunk of text what other text probabilistically comes next. This produces text, and as I keep arguing, if the probable answers are correct, then the text is probably correct, but the correctness is not inherent in the model and in fact isn’t a part of the model.
"My perspective on ChatGPT and Learning" -
I'm trying to be a better project manager for my team. This video went over how to find the "critical path" for a project simply by using data. I'm going to give this a try.
- A great long essay on why going to Mars is a bad idea (coming from someone who likes space.)
In other news...
We've been fighting sickness at home. Our three-year-old also decided he wanted to sleep in his sisters' room, so we have all four packed into one space. It's fun to watch them all growing and wanting to just be around each other, even in sleep.
I've also decided this is the year I stop biting my nails.