I don't have long-term data yet, as I just implemented this about 6 weeks ago, but last month's bill was 10% lower than the previous month and the same month the previous year. I'm looking forward to seeing the results over time.

I've been trying to think of what other automations I can try out when the price gets high. I thought maybe shorter light timers for automated lights, but all of our lights are LED, so the savings isn't really worth it.

My mom did a great job making sure I always had a computer but she didn't know enough about them to help.

Elementary school put us in front of them and taught us how to type and let us play oregon trail and print shop, but that wasn't it. And my high school has us programming in Pascal. Meh.

When I was in my 20s, in a new city, working to pay my own way I found the determination needed to teach myself. It absolutely helped that the internet was becoming this wealth of information for the self-taught. That was my launch pad. I definitely think everything else helped get me there

I had a trash-80 in the early 80s and used to copy BASIC from magazines, and then in the mid 90s I bought a book with a CD in it that taught me how to set up perl-based e-commerce store, which I set up for my mom's retail shop. Neither of those got me interested in programming.

The event in my life that got me there was around 2000, when I bought a Cold Fusion 4 book off of a street vendor - literally on Wall Street - and I spent my evenings building an intranet that replaced the Access database that my company shipped and merged across the country every Friday.

That's when I got hooked, and started building my career (and hobby).

Not sure if something like Scratch would have gotten me there (LOGO definitely didn't) - but I imagine the level of comfort I had with computers at a young age kept me in the vicinity that eventually got me there.

I see the full multi-paragraph comments in my codebases and get annoyed but also feel like the additional context helps improve the llm results over time because that history helps it know what's been tried and removed in the past. It's additional context for the system that improves with context.

The feature I want in the code tool itself (for me) is to adjust how verbose the comments are so I can read "just code", then "terse comments" then "full comments" then "full comments with historical context" (including fit commits and ticket references) and finally, full-on literate programming. And I'd like to switch between on the fly as I read through the code.

I think this is something we could actually produce with llms, and I feel the ability to switch between these modes would help the llm as well.

Sometimes I just need to see what's being done. Sometimes I need to know why. Sometimes I need to know what's been tried. Never always all of these things. And expecting to find this context in git comments doesn't feel right either.

I'm not seeing the tenses being used interchangeably with the same spelling on the site, but I may be missing it.

I love that vinyl has remained relevant. I believe that's true because of the record shop experience. It's one of the few places, besides maybe a cocktail bar, where you can expect to converse with a local expert at length while they're simply doing their job.

The movie buff at the blockbuster.

The music fiend at the local record shop

The gamer at the mom & pop game store.

The guy at Egghead Software who would let me play battle chess for hours.

Of course there are millions of them online these days, but that face-to-face interaction with the local officanado was a different thing. It wasn't a forum. And oddly, it wasn't a library either - it was a place of business - but it didn't feel like one in those moments.

Nerding out with a local stranger over a movie or a game or a band was something very different from what we do today. It was a different way for the local nerds to shine, while making a living adjacent to the things they knew so well.

One of my favorite features has been giving a recipe url and having it figure out the calories for that recipe with line-by-line entries for each ingredient. Then it keeps the extracted recipe in markdown format for the next time I want to make it.

Substitutions are also great - "I'm making [such and such recipe], but with half the sugar."

One of the others has been "I had a meal a couple weeks ago at [x restaurant], can you enter that again for right now?" Having the long-term context for entry has been excellent. And a related one, where I upload a photo of the menu, and say which menu items I'm eating and it puts together an estimate for the calorie count.

I lean heavily on a long auto-compacting claude code session for all of this. I have a UI for graphs, history, and simple entries, but claude's been my favorite means of entry by far

I quit smoking about 15 years ago, but still vape - and I remain a huge fan of the easy excuse to step away for 3-5 minutes and just zone out, or change from a large group to a smaller 2-person conversation outside.

Even back when we could smoke in bars, I would still sneak outside from time to time to settle and recharge.

As someone who hasn't had a piano lesson in about 40 years, I find myself wanting to play with the keys to match the melody. So I hear the initial melody, and then I'm practically hitting keys at random (guessing where I should be on the keyboard) until I find the first note, and then I have to listen over and over again while trying to find the second note. I kind of want to hunt and peck until I'm ready and then get tested to see if I nailed it.

Eventually I realized why and explained, "you know, I'm really just going to do a web search for what you just asked me, and maybe a couple more until I have a decent answer and then give you that answer. Let me show you how I would go about that".

From then on, they started getting into the habit of doing that for themselves. I think now with LLMs, they've kept the habit, but the LLM gives a more complete answer with fewer steps so it becomes the default. I think the magic of AI is two-fold (well, more than two, but two bullets for this conversation).

1. You don't have to "query". You can just braindump, and it'll build a context and figure out what you're looking for

2. It's conversational, so instead of filtering and tweaking results from the first query, your second "query" builds on top of the context from the first question, and you get a stronger result as the conversation continues.

https://www.youtube.com/watch?v=CHYxjpYep_M

Sometimes I daydream about a "solo mode" where the timings on lights are tighter and my music can follow me around the house when I'm up at night and nobody else is. But most times I'm trying to find the get-out-of-way averages that keep everyone happy.

Some things work great: Automated lights everywhere. Automated dimming of lights at night or sunset or whatever. Notifications when the laundry is done, or the cat litter is ready to be changed, or someone is at the door, or the garage door has been left open - all great. What music to play in what room at any time? Always changes. When to "dim all the lights" because Plex started a movie? But my son is building Legos in the dining room, and my wife is knitting and needs the couch light on. Sometimes I want it, but not every time.

For those things having a single button press is still a huge win over opening multiple apps and getting the right things set the right way for each participant.