what? on every post? is that really practical?

Yeah, this is the C++ kitchen sink approach and its the reason I don't favour rushing features. Never a second chance to get it right the first time

I'm always switching to the latest versions. Performance upgrades with no effort. Why wouldn't you (assuming you're not pinned by a dependency)

Privately yes, but at work getting the latest edge version isn't always the case

I was a volunteer in PNG at the PNG Institute of Medical Research, which at that time was run by Michael Alpers, the guy who did the initial scientific investigation into Kuru - the disease mentioned in the article. Fascinating stuff. I lived for just over half a year in Goroka the highlands, then a further year in Madang on the north coast. That was 25 years ago now but I will never forget it.

It may be tiresome, but it's true. It doesn't refute the article's premise thou, only retesting with newer models would and that would be the same tiresome requirement you already called out

I agree with you somewhat but I also just this morning read an article from a Blender educator who tried to replicate the Blender demos and couldnt get the same quality of results nor get results without errors that werent evident in Anthropic's demos

People mythologising Apple design will never get old

Expecting a corporate to police itself is, obviously, a joke

This is what I came here to say.

I'm sure the behavioural point is correct, but surfactants effect oils in a way that water does not. Water will not be as effective