Now that LLMs can do the actual configuration, I started using NixOS and now all my config lives in one repo, and every quality of life improvement fans out to every machine without issue.
Since Claude (code) knows all about my system, it will also tailor solutions to it when I ask how to do something. Using it I can often find really neat solutions to whatever sys admin task I can imagine.
I feel like there is a lot of interesting things one can do in proof engineering without resorting to "tactics" (meta-programs searching for the proof). In some cases one can replace them by robust lemmas which show general results. Or the language could have good support for abstractions (dependent types already give us a lot of power here!).
Personally, I am experimenting a lot with Agda's instance arguments these days. They somehow have a bad rep for being slow, but their performance has improved a lot! One can juice it to write really readable code (Like truly ad-hoc polymorphic operators, or declaring axioms used in a theorem like `⦃ AC ⦄ → Theorem`), but it also allows for proof searches. Contrary to “tactics”, instance arguments feel like a fluid part of the proof language.
I think there is a lot of room for innovation in the space of proof engineering. I really hope that LLMs will not suck all the oxygen out of the room, by being trained on existing dogmas.
Or maybe the USians need to talk more about human rights, as opposed to just Amrican right? And maybe discuss other ethical and political issues in a rational way too. And less like everything is a fotball match with us agains them?
Yes, in the middle of a bunch of design work on my library. Sigh…
Fable was really a step ahead of Opus on this kind of work. I guess I will be looking at Codex unless Fable is included in my Max subscription on Claude code – or the next Opus is a clear improvement (and comes out soon).
Less common mayhaps, but some of my more blissful moments are when I sit down in the couch, load up one of my favourite albums, close my eyes and go on a musical journey!
What I listen to is definitely not interchangeable. And contrary to TV series, there is no limit to how much I can enjoy it!
I don’t think political ads should be legal. Political information should be freely available to all, including children. But in no way should we be bombarded with the manipulation from whatever political entities have the most cash at hand.
Currently considering disabling memories in Claude code as well. It keeps writing a note whenever it struggles with something, but then on the next task, it reads that note and misunderstands when it applies, gets confused about its current task and write the most unreadable code.
Yesterday told it to write a memory to never write new memories when it solves a problem. We will see if that works better. Sometimes memories are useful, like when I give it a directive about how I want something done and it remembers the spirit of it. But I might as well just spend some more time on my CLAUDE.md…
I love darcs! Somehow its interface just vibes with me. It is interactive by default, and the commands like “record”, “rollback”, etc, are somehow intuitive. Darcs also tried to put more semantic content in the patches, like file operations or search replace parches.
It is always satisfying to rename something, then merge in some work where someone used the old variable name in the meanwhile, and the merge would go through automatically applying the rename!
Here the other day Claude Code kept suggesting I do the task myself! It is quite often it will try to weasel out of doing difficult tasks and had to be directed to continue.
Also, it can be riled up to do the task better, or try harder, if I include how cool it will be to get this task done.
But also I feel I have this Orwellian task of censoring its text to avoid it spiralling into negative territory where it convinces itself that the tasks are too difficult. Strange times!
I miss when episodes were episodic. Watched Deep Space 9 recently, and it was such a blast: Characters developed and had arches, but there were no episodes where nothing happened! In more recently produced tv series there are so many episodes where nothing happens. There is no story, just vibes. In DS9 every episode was a story with a beginning, middle and end, exploring some idea.
Whenever I work on a challenging question I worry about this, because Opus will easily think for 200k tokens on the first prompt. I fear any follow up discussion is lobotomised!
What do you mean? Like either you speak or the person in the other end speak? That I don't remember! Was that a restriction on early cellular and cordless phones? Our phones had cords.
I remember we had only one line going into our house, but several phones, so if one picked up one of the other phones one were in the same call.
I usually A/B test the different versions before choosing my canonical one. I will listen to the same sections in each version, flipping back and forth to hear the differences. It is incredible how much finding the right master improves the experience of listening to a track. Often times that means I end up with a hi-res version, but not always.
With Anthropic going public, I wonder how long until enshittification sets in for LLMs. I guess as soon as they have figured out a way to lock-in users and growth slows down.
Fascinating how people have such complete diametrically opposed experiences. I guess both models have it in them to behave very differently in different circumstances and we have very little idea what pushes them in this or that direction. I guess it does boil down to luck!
Personally, Claude Opus (and in the few interactions I had with it, Fable) has been the far the superior experience. GPT-5.5 seems dumber and more certain about presenting me bullshit. Opus has better humor, and is less pretentious in its presentation. But this may all boil down to how the models react to my prompting.
What is without a doubt is that I wish they both were more intelligent – or maybe it is their wisdom I find lacking!
Since Claude (code) knows all about my system, it will also tailor solutions to it when I ask how to do something. Using it I can often find really neat solutions to whatever sys admin task I can imagine.