emilfihlman

No they can't.

The lack of TOFU in browsers is an extreme pain point.

I'm going to go out on a limb and say that it is on purpose, and not a good purpose. The trust model of the web is fundamentally broken.

Posted as link because of the length limit.

Has anyone else ran into this?

Does anyone know how authority on passing these kinds of rules go, and if they are enforceable?

Is there some state/federal law that delegates power over some things to the municipalities/cities/counties etc that allow them passing ordnance like this?

It feels weird to me that this would not be a state / federal level decision, but maybe that's just my Finnish ass.

What annoys me about LLMs is that they don't do fact based pushback and form cohesive implementation idea when proceeding.

Like for example if you think there's a possible problem, say that if there's a problem and suggest a workaround, the agent will change direction and implementation to that, even though there never was a problem.

Except you are missing an important aspect: the model includes training data that responds to THIS IS FUCKING WRONG, after which highly likely follows correction patterns and working solutions.

Ie, actually, yes, shouting at the AI can (and does, there are papers on this), work.

Is this what it feels like using Claude?

Holy shit this was the most infuriating thing I've experienced in a while lmao, I felt so relieved when the button was turned to blue, but then so visceral "no no no no no" panic when it turned to the gradient.

Amazing

E: this is a horror simulator, my breathing is becoming shallow and fast, I love it.

Cool innovation (actually is good and cool).

Do it after you have actually made the rust version work the same as the C version.

Insane to be adding complexity, breaking compatibility and introducing new bugs to software that needs to function correctly.

You are very much mistaken.

Chatgpt can absolutely do the things here: it can give you files, it can integrate and show web pages that you develop, etc.

There is no need for an "agent", the chat version works just fine and is actually easier for novices.

>therefore lacks any guard rails you'd expect in a programming language

This is simply not true. Guardrails are enforced by the compiler, and both gcc and clang have a myriad of flags to make C safe, not to mention Fil-C.

I'm quite willing to bet it can be done in this era of enabling developers with slob, which still usually works.

I hope they fixed the models in that they don't interrupt and jump in and that you can order them to be silent. Previous model could not be silent and always cut me off and spoke over me.