While true to a degree, I think this is largely wrong. Wouldn't it still count as a "harness" if we provided these LLMs with full robotic control of two humanoid arms, so that it could hold a Gameboy and play the game that way? I don't think the lack of that level of human-ness takes away from the demonstration of long-context reasoning that the GPP stream showed.
Claude got stuck reasoning its way through one of the more complex puzzle areas. Gemini took a while on it also, but made it through. I don't that difference can be fully attributed up to the harnesses.
Obviously, the best thing to do would be to run a SxS in the same harness of the two models. Maybe that will happen?
Agree. Apple needs to clean up shop - MacOS has been egregiously worsening year over year. Some features like Universal Control and Continuity Camera are legitimately awesome, but they do not make up for the INSANELY slow System Settings app that gets harder to navigate with each release and which has >2s wait times for the right pane to respond to a change in the left pane. Steve Jobs would have fired the person responsible for that overhaul three years ago, it's embarrassing. Messages too needs a ground-up rewrite. Getting more elaborate emoji tapbacks doesn't make up for fundamental instability and poor syncing behavior. C'mon!