I’m still looking for valuable use-cases for LLMs. I’ve used them for low-stakes things such as a proof reader, a thesaurus1, some coding of bash scripts, HTML, and CSS (including for this website). But these are all things I can do myself. I trust the output of the LLM only because I have enough expertise and experience that I can verify the result. And if gives me a bunch of crap, I can come up with a solution myself.

My concern is that if/when I do find a valuable use-case that the big players in LLMs will enshittify their products around that same time. I’m going into the “AI revolution” with open eyes. I know that if I get any value from it, the endgame is to extract that value (and more) back from me.

So I installed Ollama on my M1 MacBook Air.

I’ve been comparing prompts of Ollama using the model Qwen (qwen3.5:4b) to the free plan of ChatGPT. So far the answers come out mostly the same, which is really encouraging.

Free tools tend to lag behind the cutting edge consumer tools, but I’m pleasantly surprised that Ollama is so close to commercial offerings. Oh, I’m sure it’s substantially worse by some measure that matters to people who are close to this stuff, but that’s not me. I don’t rely on LLMs for anything, so if I have access to a model that’s like ChatGPT from 2-3 years ago, that’s good enough for now.

Next I’d like to try Open WebUI, which is a web front-end for Ollama that gives an interface like the usual LLM products. It seems to offer some additional features and just uses Ollama as infrastructure.

1: Using LLMs as a thesaurus is by far the best use-case I’ve found. It plays to the fundamental strengths of the software because they are models of language based on large samples of writing. I can ask for a word based on feeling, tone, connotation, etc. I can say I want a word that is like (or not like) another word. As a model of words and their relationship to other words, it makes a very good thesaurus. ↩

My 2008 iMac is back in my possession after ten years of living at my sister-in-law’s house. I gave it to her in 2016 to use as a family computer where it lived in the living room and used for light web browsing. But for most of the last five years, it’s been in the basement. I was over at their house the other day and I spotted it. I asked if they were still using it (they very obviously were not, but it’s polite to ask anyway) and now I have it back.

Right now it doesn’t boot at all. That’s kind of a good place to be for me. I can tear into it in a way that I’d be too scared to otherwise. As my dad says, “It’s no love lost” if something goes wrong. I’ve already yanked off the glass front panel, which I’ve known is the way to get into these iMacs, but always seemed risky and dangerous.

If I can get this iMac working again, and that’s a big if, I’ll have to figure out what it’s still good for. My top pick is a computer for my kid. That’s something I’ve been thinking about for a while and, while I’m considering using a Surface Pro that I have sitting around, I think a desktop computer is a healthier option for a child. I’d like a computer with a fixed location rather than something that can be carried around the house, at least while early habits around computers are still forming.

Another option is to turn it into a TV or entertainment center. Half the reason I got the iMac instead of an upgraded Mac mini (what I had at the time) or a MacBook (what everybody else had) is that the screen and built-in speakers made it great for watching movies and listening to music. Also Macs at the time had an infrared remote (which I’ve lost) that controlled Front Row, a 10-foot user interface in Mac OS X for viewing on the couch (or, God help, futon). It was a perfect machine when I had a big MP3 library and Netflix’s 3-disc-at-a-time plan. I could re-create some of that by installing Kodi/LibreELEC.

It’s too underpowered to be competitive as a generic Linux/Debian environment for me. There’s no reason I’d ever choose it over another device for that purpose. But maybe I could use it as a terminal-only writer deck. Foregoing any graphic user interface is the best way to breathe life into old hardware. Most of the RAM and CPU usage comes from just displaying things on screen. Without a graphical interface, entire categories of resource-hungry applications just aren’t viable like web browsing or watching videos. And multi-tasking becomes a less intuitive, if not completely impractical, thing to do. Going terminal-only is always an option, though it’s a change that comes with very steep trade-offs. I don’t look forward to explaining to my spouse that this bulky computer taking up precious space can’t do any of the things that we understand a computer to do.

Anyway, I’ll need to get this thing cleaned up and diagnose the boot issue before I can do anything. If nothing else, it’s interesting to think about uses for older Intel-based Apple computers.

I’ve followed Charlie Warzel’s Galaxy Brain since the height of the pandemic, even before it was aqcuired by The Atlantic. I’ve nodded along for years as Warzel wrote about internet bullshit, but lately the newsletter has gotten weird even for me.

Charlie Warzel, The Feeling of Control Slipping Away:

Culturally, the flood of slop, AI influencers, fake accounts, and AI tools is blurring the lines of an already post-truth age.

I can’t make sense of that!

Is an “AI influencer” an influencer whose area of influence is “AI”? Are they an approximation of a human influencer created using generative tools? Flood of slop, fake accounts. Where? Which AI tools?

I give myself some credit here. If Galaxy Brain is now unrelatable, that’s because I’ve insulated myself from internet bullshit. But I did it, in part, because of the influence of reading Galaxy Brain. I feel I owe something to Charlie Warzel. It’s concerning that his brain might be getting eaten by the subject he studies. So please send somebody to check on him. I’d hate to lose a good one to internet madness.