[scroot]

Excel is programming a computer. So was Hypercard. Any UX that reduces user "savvy" rather than enhacing it can hardly be said to be improved. A large part of the developer community has a very static idea of what a computer can be or what programming can be. What does it mean to author in the medium and why have we stopped trying to figure it out?

[regularfry]

I think it depends what you mean by "savvy". A UX which reduces what the user has to think about to get stuff done is generally positive. That's what an abstraction is, and we intentionally create them all the time.

The counter-example is trite, but it's true: it's a very good thing that I don't need to know how a Xeon is going to reorder the instructions that v8 is going to turn my webpage into after Babel has turned my es6+whateverextensions into something node can actually execute. I can go down the stack and find out if I absolutely have to, but it's a total waste of time otherwise.

I also don't think we've stopped trying to figure it out. We see new languages, new environments coming forward fairly regularly. My instinct is that the reason it seems that way is because the computing field selected for people interested in that stuff early, and more recent incomers are a) people who don't yet have the experience to understand where the limits are; and b) as a population are less interested on the whole in asking those questions, because if they had been interested, they'd already be present. It's also dramatically harder for a single project to become ubiquitous the way Hypercard was, simply because of the size of both computer-using and software project populations. It's a statistical artefact, in other words.