- China is the largest open weight provider, with Mistral and Cohere delivering a few other models. There isn’t much else internationally
- (I think OP is suggesting) this would effectively ban Chinese models in the US, which would be an interesting case. Who knows if they could have theirs reviewed, or if we’ll see another FCC approved router situation.
- that Chinese models are censored is a very common criticism. If American models are also censored that looks bad.
- this will be awful for self hosters and local inference. Imagine if HuggingFace had to drop non-American model weights. That would effectively kill them.
> - that Chinese models are censored is a very common criticism. If American models are also censored that looks bad.
It's even worse than that for American models.
As an American, if I want to run a model locally and have to choose a censored model I will choose a Chinese censored model over an American censored model especially if it is the Trump administration doing the American censoring.
Chinese censorship is mostly directed at things that would not reduce the usefulness of the model for my applications. I doubt that would be the case with Trump censorship.
Same for products that spy on me. If a car for example is sending my travel log to Korea or the EU or China it is annoying but none of them are realistically going to do anything with the data that would seriously harm me. The risk is orders of magnitude higher if US governments or US law enforcement gets it.
I feel like it’s just as likely that China and MAGA mutually censor each others sensitive answers. Who in MAGA benefits from a Tiananmen focus? Who under Xi benefits from calling Jan 6 a failed coup?
You can make bombs out of pressure cookers. Since they are so good at holding pressure, you seal off the over-pressure release valves and then pressurize them until they burst (usually via some stupid or illegal means to begin with) then when they burst there is a ton of excess pressure. Big pressure = big boom.
But if you're not working with cutting torches and welding equipment, there's no easy way to open a propane tank enough to put an explosive in it (assuming purely amateur equipment) and then seal it up again good enough to hold high pressure. A pressure cooker on the other hand is ready made to have a big opening to put lots of stuff inside and then seal it up again.
Your comment unironically is something I prefer and one of my biggest pain points with Linux.
As a newb, I'm sure there's something called with a mycommonproblemd name that has a stateful interface. But sometimes that all adds up to make things feel complex. And it let's me make stupid mistakes, like I forgot to close or open a port on firewalld, or I disabled a container but forgot to commit a change to my systemd units.
It's nice to just have a nice file called myservice.nix that tracks the firewall port, name, systemd startup and update scripts.
I’ve always struggled with what should be in Claude.md that doesn’t belong in readme.md or a similar supporting file.
I tend to include a well documented justfile, so between the readme and that common commands are covered. If there’s a style guide it should be its own file, or summarized in the readme.
If Claude is making errors I tend to just update my global Claude file, but I haven’t updated it in 6 months — only to disable Claude signatures on generated commit messages.
README.md should be for humans CLAUDE.md for machines.
Most agents use README as a storage for EVERYTHING related to the project by default, which is annoying for humans who just want to figure out a) what the thing does b) how to install it. Then you start reading and there's some intricate documentation on how data flows through the application etc.
If you're only working on your personal projects with no collaborators, just a global claude file is just fine. Per-project files are more for things that are specific to that project.
I’ve got a NixOS Qemu VM I use to run openclaw in. I had Claude help me set it up, and it runs local models on my own machine in a config based sandbox.
Why should Claude block or charge extra to work on that?
Why should Claude care if I have instructions for Hermes or OpenClaw in my project repos?
This fingerprinting is incredibly sloppy for how much access to a machine Claude code has.
Now you've learned the advantage of knowing how to do things yourself. When you depend on untrustworthy agents, you shackle yourself to their idiotic whims. Be careful who you partner with.
I run an OpenClaw VM and used Claude Code to build the VM scripts. The VM is connected to local llama.cpp, so OpenClaw and the models are running on my own physical hardware.
I’m actually trying to move back from the Claude Code style, I feel like it’s easy to become distant from your own code, and I am feeling uncomfortable with that.
I’ve “vibe-coded” some projects and when I start to find issues or go to refactor them I don’t have that memory of why decisions were made, because many decisions were never made.
- China is the largest open weight provider, with Mistral and Cohere delivering a few other models. There isn’t much else internationally
- (I think OP is suggesting) this would effectively ban Chinese models in the US, which would be an interesting case. Who knows if they could have theirs reviewed, or if we’ll see another FCC approved router situation.
- that Chinese models are censored is a very common criticism. If American models are also censored that looks bad.
- this will be awful for self hosters and local inference. Imagine if HuggingFace had to drop non-American model weights. That would effectively kill them.
reply