21 karma · joined March 20, 2025
If you look at the terminal-bench@2.0 leaderboard, you'll quickly see it's actually one of the weakest agentic harnesses. Anthropic's own models score lower with Claude Code than with virtually any other harness.
So it's quite the opposite. Claude Code is arguably the worst harness to run models with.
What makes this worse is that Apple's refusal to expose any public APIs to control workspace behavior so you can't even work around their shitty choices.
Instead of iterating on existing functionality, they launch flashy additions like Stage Manager only to abandon them immediately.
b) He is an investor in helium
> The desktop metaphor is still unmatched when it comes to productivity
Well, duh? Of course it dominates productivity when we've spent fifty years building every tool, workflow, and interface around it. It's better than anything else, because there isn't anything else! It's like saying horses were the best transportation in 1800 because all the roads were designed for them.
The article also sets up a false tablet vs. desktop dichotomy and overlooks that tablet OSs inherit the same PARC-era WIMP assumptions.
For a more thorough critique of the desktop paradigm, I'd recommend the Liber Indigo (https://youtu.be/As_SiWqC5tc?si=_rKuZQN22ZkbqGvW) video series on youtube, which does a stellar job highlighting these foundational problems.
Having a database isn’t mutually exclusive with the core functionality. You can simply not use it.