I'd have mined the copied libraries with something that makes it possible to later change terms and extract fees, as it'd be expected that nobody reads the terms for such service

I'll never touch any git wrapper, because they've lied to me before and I can use git already. Everything that was there to be sped up has already been made into zsh functions.

Nvidia can also say no, they won't have choice but yield or not have AI at all

It reads like a joke without a punchline

Maybe, but will they have to fight with borrow checker for doing some other than (the very OOP) DOM components? They'll obviously use both for a long time in the future, so more functional places can get Rust, while more OOP places can benefit from C++

Do you seriously think LLMs will not just spam unsafe blocks in it like they do with any task ever?

I'd generally be quite surprised to see LLMs spam unsafe blocks, both because that's behavior that I haven't observed while using them and because that contradicts my mental model of them where they imitate the styles of code that they were trained on (which in rust generally does not include spamming unsafe).

You do want a browser with RCE, but you want it to keep the it sandboxed. The hard part is executing the code safely

I have my doubts it'll ever be "finished". Servo gives strong vibes of a project that will avoid performance hacks, because they're not nice/state of the art code. I have no evidence, it's just the energy I've picked up from it

None at all, the generated AST and bytecode are stated to be identical