Busy day today, not much blogging probably. #
Monday, May 26, 2025
I do find all the LLMs stuff interesting, and I've noticed the latest Anthropic release for Claude seems to be reverberating with folks, especially for its better coding ability. The developer in me was immediately curious about this Simon Willison deep dive into the system prompts. Though there was definitely some interesting stuff in there I found myself skimming through it, because like all things LLM, it's all so wordy, and you also end up feeling like some sort of computer psycologist. I wonder whether my web developer trained brain will ever get as into this as Simon appears to be.
At the minute, to me it feels a bit too close to horoscopes than I am comfortable with, no offense to any psychics and mediums that might be reading this post. I find myself craving for the unambiguity of if statements, for loops, variables and of course async/await. #
The
Since moving to Android I've been using
Quite a crazy few days for me, hopefully things have settled at least for a short while. The world keeps on turning :)
It's interesting to see that
I just rewatched the very cool
Netlify support did finally get around to
All sorts of strange issues today. As well as all the deploy problems, I had some really quite bizare interactions with GPTs today. I have previously had some good experiences working on CSS with AIs, but today was with javascript, and it was really quite terrible. I was trying to do something very simple, and the AI was repeatedly failing, and adding complexity while doing so. So I asked it to remove the styling that it had added without asking me, and it said yep removed it. And it hadn't. This went on for at least 30 mins. Over and over it would say that it had removed it, but then it still hadn't, and it would even check itself and say, oh yes I didn't remove it, then still fail to remove it.
What with the Trump tarrifs, new US trade deals being struck, and the new pope, it feels like a lot of big things are sliding into place. I wonder whether this could be the beginning of a nice bit of momentum for everyone.
This whole transition to using AI tools is going to be very complex. Of course that's nothing new to hear, but I get a sense that some are starting to get a feel for the size of the issue at hand. Reality is very very complicated and it took humanity literally millions of years in a shared reality to build the current non-AI "solution". I feel like we need to be careful, but also not too careful, sometimes you have to press on reguardless. But also, empathy, kindness.
Figured I'd try for one more feature for