tym0

Is it really that different in implementation? Go's yield points are implicit but aside from the devex side how is that different from async await?

Wait groups do use concurency primitives underneath. My point is that the primitives that go provides are too easy to use incorrectly. And the language makes it hard to build nice abstraction on top of them. So I generally steer people away from using them unless strictly necessary, channels in particular.

We're heavy user of Go at work. Go also makes it way to easy to write bad concurrent code and hard to write good one.

Stick to err/wait group and go routines and it's OK. Any PR with a channel or mutex I'll assume the author made a mistake.

How are things like that, that were preserved (I'm assuming) by the dry sand they were buried under, then gets preserved by archaeologist for the future?

A lot of pigments react to oxygen, and the sandy winds must not be kind to this kind stuff so I'm curious.

I know they've had to close Lascaux to the public for example.

Don't really see why. The issue with PostmarketOS is that there isn't enough people working on it. Not that the phones are locked down. Otherwise phones with open bootloader would have good support.

That looks nice but no Xbox support.

As other have said the apps for devices is still the issue with Jellyfin.

For TV you pretty much have to use Kodi and all the decent theme require a surprising amount of other plug-ins to work...

Anyone know of a decent theme for Kodi that won't fall over next time it updates? I just need "Resume Watching", "Newly Added" and alphabetical lists.

Yeah adding tools dynamically seems good but removal is a terrible idea. Instant cache nuke.

I also really don't see how the hacky command interceptor is that much better than a dedicated tool. You're one which away from a very confused LLM.

Yeah I've done more research and that's what I meant in the edit you replied to.

But that's not the full reasoning token context, just a snapshot of the latent state at the end of it, no?

Have a look at the gemini ones they're pretty small.

Sorry, I've realised I was only partially correct.

Gemini[0] for example passes along a snapshot of the reasoning state but it's not the equivalent to keeping all the reasoning tokens in the context.

[0] https://ai.google.dev/gemini-api/docs/thinking#signatures

Edit: Apparently it does take the same space in the LLM latent space so I was wrong.

I mean the thinking does not bloat the context window because it gets dropped at the next request.

That's not my understanding of how most agents work. This is what a chain of request/response looks like:

  Your Prompt 1: Prompt Content 1 -> cache-1
  LLM Response 1: <Thinking>Thinking Content 1</Thinking> Response Content 1
  Your Prompt 2 (client side): prompt-1 + response-without-thinking-1 + Prompt Content 2
  Your Prompt 2 (server side): cache-1  + response-without-thinking-1 + Prompt Content 2 -> cache-2
  LLM Response 2: <Thinking>Thinking Content 2</Thinking> Response Content 2
  Etc...

So reasoning gets dropped from context and you still get cache from the accumulating requests.

Edit:

I've realised I was incorrect, the thinking doesn't get passed back and forth but the latent snapshot does which result in using memory just the same.

> does not bloat up the context window too fast with reasoning tokens

How much does that matter if it's reset at every turn?

All agents I know of have a hook system that can call arbitrary scripts.

Yeah the compact form factor and energy efficiency were the strongest point for me.

Yeah, unfortunately the TV app (Xbox for me) offering is really poor.

I've been playing with Kodi with the Jellyfin plug-in but it requires such brittle tower of themes and plug-ins to look like Plex that I've yet to make the switch.

The homescreen on the Xbox app is now full of "Apple TV", Trailers, free TV and the like. Enough of it that you can't see you own stuff on the homescreen.

Last time I searched I couldn't turn it off. But maybe I missed that setting.

I've used the Void userspace for a while as my WSL layer and I do really like the package manager.

Anyone actually running Void on real hardware? What's runit like?

I think as long as it's just an installer with an extra repo and the rest keeps pointing to the Void repos then that sounds manageable.

There's about 20 packages in their forgejo instance.

Doesn't mean I would pick it up but if you're maintaining that for yourself and your mates that's fine.

Honestly it was quite a whiplash to go from what looked like a good article to something that seemed completely made up. But I would chalk that up more to my reading comprehension than your writing.

Thank you for spelling it out for me because I thought I was looking at a completely hallucinated AI article...

Right, are any of the things mentioned in the GP really required if you only want this init?

I know they've attached all these projects to the systemd brand because they thought it would beneficial but it's hard not to wonder if we could avoid all those discussions if the umbrella project was called something different...

Thanks I'll give it a shot when I have a free weekend.

I think I may have complicated things by using linuxserver.io images which already require some id mapping trickery on the rootless podman side as they expect to run as root...

I did have a terrible time setting up rootless podman in a proxmox VM with virtiofs.

I just couldn't get the ids to work. But I've not tried the specific thing they do here.

Ooh, I need to look into that id mapping technique.

I attempted to port my homelab quadlet setup to a disposable VM on proxmox with the state folder mounted via virtiofs and after losing a weekend I realised that it essentially wasn't compatible with rootless podman id mapping and I gave up.

But maybe I missed something...

I feel like this is really due to the harness.

Gemini CLI at work has the same issue: it'll prefer hacking your workstation over just asking you how to proceed.

I think the harnesses are setup to have a bias to action otherwise the LLM would just stop all the time when doing trivial task but it also mean they'll keep going when the "obvious" path is to just prompt the user.

They used to have their own "youtube script interpreter" that was kind of fascinating. But yeah as you said they switched to proper js runtimes recently.

For me, and despite the constant enshitification of Plex, the apps are still better. But I'm really close to the tipping point, I've been tinkering with making Kodi not suck but custom themes really seem like piles of unsuported plugins.

What other countries do? A chemical test on the field and a more accurate one when they get to the police station.

Not sure why this is so down voted. I have a touch bar era Intel Mac, I regularly check the state of Linux on it and they are still somewhat poorly supported. Some drivers are out of tree, audio and sleep doesn't necessarily work, etc...