> A lot of the things people are making now are being driven by young people. Young people have an ignorance about the past, so miss some of the lessons, but th

> A lot of the things people are making now are being driven by young people. Young people have an ignorance about the past, so miss some of the lessons, but they also lack the cynicism which prevents revisiting things that people have determined "don't work" in the past. Indeed, much of the progress in technology today is coming from revisiting stuff that didn't work out in the past but has since become viable (much of modern ML, for example). Conversely, it seems like the actually novel stuff coming out today "doesn't work" out-of-the-box at a much higher rate than when new stuff was introduced in past decades. I'm seeing pervasive problems with new technology for use cases that "just worked" consistently with the incumbent tech right from the outset in the '90s or earlier. There has been some progress in consumerizing interfaces, but this has been accompanied by drastic regression in functional consistency and long-term reliability under the hood. And while UIs that only expose designers' preconceived use cases might be a bit more straightforward for end-users, accessibility to the underlying functionality of technology as technology has vastly regressed. It feels like Sturgeon's law has gone from 90% to 99% over the past 10-15 years or so.