What if instead we respected state sovereignty and international law for the betterment of all instead of continually stirring up shit to benefit of a few powerful figures?

Because as a man myself I honestly wouldn't be able to say which privileges I've lost that my forefathers enjoyed, besides sexism with impunity. In fact, I have it easier, for the time being at least. No military conscription for one. And with the recognition that the patriarchy hurts all I've been able to actualize myself in a way that is more authentic to myself than the constraints of past generations would have allowed.

(/rant. Pickup trucks have a purpose, such as in low density rural areas. That’s not the case in Belgium, for instance. Either you need a van because it’s for work, or you rent a lightweight van on the few occasions you actually need to transport a lot of stuff.)

So it’s certainly not perfect. An isolated VM or a VPS provides the best guarantees. For me though it’s good enough. I’ve put my risk profile at: ‘don’t fuck up my system directly and don’t exfiltrate secrets directly’

A good example of why is project-local .venv/ directories, which are the default with uv. With Lima, what happens is that macOS package builds get mounted into a Linux system, with potential incompatibility issues. Run uv sync inside the VM and now things are invalid on the macOS side. I wasn't able to find a way to mount the CWD except for certain subdirectories.

Another example is network filtering. Lima (understandably) doesn't offer anything here. You can set up a firewall inside the VM, but there's no guarantee your agent won't find a way to touch those rules. You can set it up outside the VM, but then you're also proxying through a MITM.

So, for the use case of running Claude Code in --dangerously-skip-permissions mode, Lima is more hassle than Nono

Good DX, straightforward permissions system, starts up instantly. Just remember to disable CC’s auto-updater if that’s what you’re using. My sandbox ranking: nono > lima > containers.

Because that could be easily resolved by factoring % cache hits into the usage limits.

> Literally everyone does this.

Never a strong justification, much as I like Anthropic in general.

Why is the 'Mercedes gas station' selling gas 85% cheaper but only to Mercedes drivers?

Why is the 'Apple electric company' selling cheaper electricity to households with Apple devices?

They're not the strongest analogies, I'll admit, but that's what it smells like to me.

That said, some good can come of this in the sense that it will (hopefully) discourage these kind of schemes. They don't create value and they harm smaller competitors, who now need to divert resources to increasingly sophisticated bot detection.

But I don't see how it's fraud in the criminal sense. That's just my judgement as a citizen, not a lawyer. All I see is the shopping mall shaping criminal law to its own benefit.

As for the military bases, yeah, stay away from those, kids.

I see no victims other than large streaming services who failed to account for a changing reality.

I’m getting ‘because of torrenting metallica won’t be able to afford its third private jet’ vibes from this

Even as a CC user I’m glad someone is forcing the discussion.

My prediction: within two years ‘model neutrality’ will be a topic of debate. Creating lock-in through discount pricing is anti-competitive. The model provider is the ISP; the tool, the website.

I mourn the innocents on both sides

I consider it a real loss. When designing commands/skills/rules, it’s become a lot harder to verify whether the model is ‘reasoning’ about them as intended. (Scare quotes because thinking traces are more the model talking to itself, so it is possible to still see disconnects between thinking and assistant response.)

Anyway, please upvote one of the several issues on GH asking for thinking to be reinstated!

I’ll take a pay cut any day for the ethos of the EU.

  Palantir [...]