Yes. Not only that, but by observing the kids they iterated on different versions of Smalltalk. The approach there was quite different from what you see today.

Yes. Not only that, but by observing the kids they iterated on different versions of Smalltalk. The approach there was quite different from what you see today. They built upon actual psychology and spent the money and time conducting studies. From that we got things like GUI, OO, etc. When you see Kay harping on about the dismal state of affairs today, I think it's that he sees this constant pedantic bickering about static vs dynamic typing and all the rest as meddling around in the complexity of things that should be irrelevant. To the point of the original post, it's also worth noting that Smalltalks were not just text environments, and that while yes much of the code was written in text, these kids were not writing text files that they then compiled/interpreted using their PDP Terminal emulation programs. They were doing everything in the live environment itself. Considering the existence of things like Smalltalk and Hypercard, whenever some uber programming expert tells you this-or-that reason we cannot have such systems in wide use today, it's akin to saying: "See all this great stuff we used to have? We can't have it anymore"