zavec

I've been getting into emacs somewhat recently and really enjoying it so far, though I do worry a bit about how much it can do and whether it will lead to any sort of "oops, you opened a project that had a thing that autoexecs or exploits a parser and now you're pwned". I know that's not a unique problem to emacs, but it feels like the enormous amount of things emacs can do gives it a bigger attack surface for that (and I don't know my way around it well enough yet to be confident that I've looked at all the places that might happen)

Maybe the solution is that I should just be running the entire thing in a sandbox in any case.

I'll be very interested to read any followups about the part you described at the end about getting it to do more "teaching," both from the perspective of how we can use this to help newer devs out but also from the perspective of helping ourselves understand the software being built.

One thing I've been worrying about recently is the notion of comprehension debt and making sure the humans can still understand the system (so as to be able to do incident response or something if the LLMs are down). The slower, more methodical workflow described here should definitely help by allowing checks like proper docs whenever a piece of the architecture changes for instance, but I'm curious what else could be done to help the humans understand as much of the implications of what the AI is doing as possible.

Edit: having the code author and reviewer be completely separate like you described below is also probably pretty helpful for making sure changes are comprehensible from the outside.

Yeah I don't entirely follow why having had it generate an integration test you'd just throw it away afterwards.

Very cool! I love reading about details like this.

I'll definitely check both of these out! I've been working on a design doc with my assumptions, threat model, and required security controls for an agent sandbox so that I can use it as a baseline when comparing different offerings.

Just a few days ago there was an article about how they lost the bluebird logo and the word "tweet," so it seems possible!

Between this and the nix/guix stuff, Farid is definitely becoming one of my favourite mad computer scientist bloggers, right up there with Justine.

(Or computer mad scientist? Whatever (mad scientist) && (computer scientist) is)

The way I've done it has been when there's a longer task, I give it a markdown document that has a plan with numbered steps, and then for each step start a fresh context window and tell it to update that doc as it goes with any decisions taken or deviations from the plan, or other context needed for future steps. If it gets close to the end of the context I tell it to summarize the current state and anything a fresh context would need to know.

I always wonder about the choice of hjkl, because don't people normally rest their hands one row to the left, on jkl;?

I was just looking at conductor and was not very jazzed about the fact that it was running the agents directly on the host. Being able to launch from a terminal means this one can (hopefully) run from inside the sandboxing setup I use for coding agents.

I think generally harnesses have this. In claude code it's `"preferredNotifChannel": "terminal_bell"` in the settings.json, pi and opencode looks like you have to either add a hook yourself somewhere or use an extension.

Hmm, where have I seen a story before where a kid wants their parents to read the same book over and over. I can't quite put my finger on it...

It definitely seems brutal, but when I stop and think about it it's actually probably more humane for the prey than getting eaten alive, right?

Hell, I grew up with English as my first language and I still learned a lot of obscure vocabulary from magic!

It's definitely not so good these days, but a format called premodern is getting more popular. You may find it worth looking in to that one.

Homestuck is finally happening?

Alive 2007 is a bit late to be the 90s though

Lock picking is great! Many security conferences will have a "lockpick village" where there are just a bunch of locks, lockpicks, and volunteers teaching people how to do it.

This looks really promising, I am curious about the choice to use containers as the isolation layer though. If the goal is to treat agents as untrusted and isolate them fully I feel like microVMs would be a better option.

If it supports OCI runtimes though then maybe kata containers can be plugged in, I'll have to dig in after work and see.

> The courts decided that rap records had to clear every single sample, thereby basically destroying the art form, but now you can literally feed every book into a blender, piece another book together out of the pieces, and sell it.

Are they still enforcing the old way on hip hop samples, or has that changed with the recent rulings? If the new way of doing things is applied fairly to everyone that seems like a win.

This sounds fascinating, if you ever write more about it I'd love to read it!

Last rc is 2025 but there are at least still new commits (one from 3 hours ago)

> IMO the issue is not propaganda at all, but real physical problems that are not being addressed.

I don't think those are mutually exclusive. There can be real problems, and propaganda can magnify those and lead people to decisions that are for the benefit of the propagandist rather than things that will actually solve the problem.

It was the same when I graduated 6 years ago. We had projects to test our ability to use tools and such, and I guess in that context LLMs might be a concern. But exams were pencil and paper only.

There's also a sex toy

You haven't been getting the checks? Bring that up at our next secret cabal meeting.

I think you can with a user.js file, unless they changed that?

It's bitter for me because I like looking at how things work under the hood and that's much less satisfying when it's "a bunch of stats and linear algebra that just happens to work"

I guess it's nice for non-technical people who don't know how to use `about:config` but beyond that I don't really see the need. Hopefully adding that extra layer of indirection doesn't mean the users will have to wait too long for security patches.

Pragmatic is not bothering to switch to the shiny new thing just because it's shiny and new if your current tool doesn't give you any problems.

This is exactly why I just bought one! I take the bus to work, but I live about 1.5k from the nearest bus stop and it's pretty flat.

Unfortunately it's just starting to be icy here so I probably won't be able to make much use of it for several months, but at least it's something to look forward to!