What Could Possibly Go Wrong?

I just finished reading Mathematica: A Secret World of Intuition and Curiosity by David Bessis. You should read it. It has some great ideas about learning. It is a pretty quick read too.

I'm thinking more about thinking at the moment because of it. This morning, I thought about how AI can be used to help you with computer stuff you don't understand. However, I read a story recently that said that a lot of young programmers don't actually know how to program. They are over-reliant on AI. And it occurred to me that AI is a prosthetic. Don't have a skill or ability? Strap it on, and it can (more or less, depending on the quality of the AI) bring you up to an average level of competence. This works well for skills you don't want or need to develop. My interest in gardening is enough to want to plant stuff and watch it grow, but I'm not interested enough to go to university to study botany. I'll learn some by doing, but I'll never be an expert. Using AI also is great for people who lack something that they cannot acquire. Yesterday, I was freaking out, and I used the AI to hold my hand getting me through my day, as a kind of prosthetic functionality. When you get to skills that you want to learn, it can also be a prosthetic, for better or for worse. I use the AI to write emails in Portuguese. If I wanted to learn Portuguese faster, I would write the emails myself and then use the AI to correct it. However, at the moment, I am more interested in getting things done and saving my mental resources rather than learning faster. However, I have used the AI as a prosthetic teacher ("explain the difference between x and y") and as a prosthetic study partner ("let's practice speaking Portuguese together"). For skills you already know, it can be like a prosthetic assistant. Obviously, if you use a prosthetic in place of learning something yourself, you will never learn it. I think it is easier to understand why if you think about it in this way.

I finished up the painting with the path that I talked about. I'll hold off on posting a picture here until the recipient gets it. I've also decided to simplify the painting I was going to do for my uncle. I have a lot of work to do, and there is no reason to make something super complicated.

In other news, I started working on getting Claude to write me a story that will teach me statistics. The first chapter didn't have any statistics in it, but in the second revision, it was a perfectly acceptable chapter. It mainly had the problems that lots of writers have. I asked it to not use any tags except "said", "asked", or "replied" with dialogue and to not use any adverbs with dialogue tags. I pointed out that it was telling much more than showing ("Anna felt excited" instead of "Anna leaned forward, eyes fixed on the screen") and that there was almost no sense of place. The story was a bit rushed too. When it fixed those things, it was pretty good. I'm looking forward to the actual statistics.

Speaking of Claude, I asked it to help me organize myself, since I'm still having trouble getting things done. The biggest thing I've noticed is that it has me do all the quick, not terribly important things first. A lot of those things, like vacuuming, aren't a huge deal, but having a clean house is nice. Since they take little time, I feel like I've gotten a ton done and am not stressing at the end of the night over all the little things I should have done.

Like many people on the internet, I have spent some time using ChatGPT, an AI chat program that is good at a lot of things. For example, I asked it if I should use woven or t-shirt material for a braided rug, and it said that I should use woven, because it is sturdier. Anyway, I've been wondering if I could make interactive fiction, like choose your own adventure, with it. I looked it up, but I didn't get much information. Then, today, it occurred to me that I could just ask the program, so I did. It was very enthusiastic about it. I'm going to hold off until the next release when there is a chat log to work on it. However, I think this could be a lot of fun. Also, I can say that the new version is much more friendly than the previous version. That version was kind of grumpy and snippy.

I think I mentioned yesterday that I finished putting texture paste on everything for the hutch, so now I'm working on the art. I've had to change things around a little bit, as a couple of the designs that I planned on using were just too big for the space available. With that done, I started working on doing test drawings. The first one, a symbol from a video game, wasn't too hard. I'm glad I did it, though, as I think when I do it next time, I'll do a better job. The hard one, though, is the sphinxes from The Neverending Story. In my reference photo, one of them is just a little bigger than the other, which I didn't realize when I was planning it. I'll have to make some decisions about how to deal with that soon. I'm looking forward to it, though. I am sure that when I'm done with it, I'll be much more ready to paint the drawing for the hutch. Also, this is much more fun than texture paste.

Speaking of finishing up things, I finished the hoodie! As I may have said before, the sleeves are a bit like a suit jacket, which makes the hoodie look kind of frumpy. It is baggy at the waist, and the diagonal design across the chest is not attractive on me. However, now that it is all together, it looks pretty good. It is also very comfortable. I'm not sure how I feel about the thumb holes. The actual thumb hole design is great. However, I may not actually like them. I don't have to use them, of course. If I don't feel like I need them, I may leave them off the next hoodie I make.

Oh, and in boring news, it is supposed to just pour rain over the next 24 hours, so probably our roof will leak. And I still can't hear out of my right ear.