wild_egg

On the mobile point, there is https://github.com/instawork/hyperview

Your Go server can have endpoints that render XML instead of HTML and basically get the same server-driven experience of your HTMX site. Fully skips the need for the app review process since you're not updating the actual client app code to make UI changes.

They're definitely rare. Mirror's Edge is almost 20 years old. Reaching back that far for an example just reinforces how rare it is.

If you tally all the FPS releases in a given year, a supermajority are going to have male protagonists.

How is neurosymbolic not aligned with the bitter lesson? The bitter lesson is completely agnostic to architecture.

Why on earth would you spend any time at all convincing an agent of anything? You say "just do it" and off it goes.

No it doesn't.

I'm a divided individual.

I've spent an obscene number of hours learning how to get reliably good quality code out of these things. I'm actually very happy with where the tech is right now and can't imagine ever going back to typing code by hand.

But I absolutely hate how companies and society at large are acting because of this stuff. It feels like all rationality has flown out the window. So I'm just staying in my sandbox with my little toys and hoping the mass psychosis blows over at some point.

Which is what makes putting an LLM inside a lisp so much fun

FilePilot has been a thing for a while now

> why are you comparing Claude Code to your editor?

Because the editor does more. All the compute-intensive parts of the agent are in the cloud. Zero reason for an agent harness to require anything beyond a potato to run.

What does a breakthrough look like?

Deferred tool loading is not part of MCP. It's a Claude API special parameter that most other LLM APIs do not support.

There's a VS Code extension that got me through the Common Lisp learning curve using my familiar tools

https://marketplace.visualstudio.com/items?itemName=rheller....

We train dogs to be subservient but that doesn't automatically mean we anthropomorphize them

For the real emacs experience you could use this mod to render an IDE in Minecraft editing the mod that renders the IDE.

When you say modules, is that @scope or something else? I can't find any reference to a native thing called modules but this seems to fill the same role.

That still leaves the question of how one gets their foot in the door. Lots of us are aware of the budgets but we don't get how's sales work at that level.

> $32 / 1M audio input tokens ($0.40 for cached input tokens)

Anyone know how much audio is 1M tokens? I have no way of knowing if this is fine or prohibitively expensive.

Definitely not 50%. Aside from the aforementioned anticheat titles, my entire Steam library works flawlessly and has for several years now.

I'm sure there's some long tail of titles with poor support, but how many people are actually playing those?

Unless your games require kernel anticheat, there's a decent chance they actually run better on Linux than Windows these days.

Can get a GUI with the same prompt if you tell it to use TCL/tk instead of Go + Charm stuff

Skill issue. Literally. Make a SKILL.md that has the agent leverage subagents to do all work. An implementor agent does the thing, and then a separate agent reviews and verifies afterwards. The fresh context window of the second agent doesn't have the shortcut chain of thought in it and so it will very happily flag if the first agent cheated. Main agent can then have a new set of agents go fix it.

This has completely solved the cheating and fudging to make tests pass for me.

Being able to fire someone, thus causing potentially significant hardship, is considered quite normal and acceptable in most lines of work.

Hey, so... Thanks for this. I've been building ticket systems and agents and whatever else as flat files in git repos lately and now I see I have to extend that to actually managing the repos themselves.

This is going to be so nice.

If you know lisp, just reach for Coalton instead of Haskell

Technically, I think this is meant to develop Coalton, which is also statically typed and incredibly effective as a language for agents. All those ergonomic benefits that humans enjoy also allow AIs to develop lisp systems quite rapidly and robustly.

At the core, they're really very simple [1]. Run LLM API calls in a loop with some tools.

From there, you can get much fancier with any aspect of it that interests you. Here's one in Bash [2] that is fully extensible at runtime through dynamic discovery of plugins/hooks.

[1] https://ampcode.com/notes/how-to-build-an-agent

[2] https://github.com/wedow/harness

Inference, in and of itself, can't be completely unprofitable. Unless you're purely talking about Anthropic?

But

> If you want LLMs to continue to be offered we have to get to a point where the providers are taking in more money than they are spending hosting them

Suggests you just mean in general, as a category, every provider is taking a loss. That seems implausible. Every provider on OpenRouter is giving away inference at a loss? For what purpose?

A week of downtime every decade I think still works out to a higher uptime than I've been getting from parts of GitHub lately. So I'd consider that a win.

Where did you see a haiku comparison? Haiku 4.5 was my daily driver for a month or so before Opus 4.5 dropped and would be unreasonably happy if a local model can give me similar capability

That's really not remotely the same thing