This is a great insight, but there's a third angle too: a lot of these 80s tools assumed you'd invest in them, and not just in cash: you'd give them a lot of ti

This is a great insight, but there's a third angle too: a lot of these 80s tools assumed you'd invest in them, and not just in cash: you'd give them a lot of time and attention and learning and hence reap expert-user rewards. The world simply isn't like that now, except in niches. And if you're niche, you can't establish a new file format in a networked world, so you end up having to be interoperable, with markdown in text or xls for spreadsheets. And if you're interoperable, you end up shrinking your market even further: people your users are sharing with don't need to use you. You really can't build much of a moat on "powerful UX that boils down to a sharable file format". The only way to make space for yourself then is to become a platform in your own right (see emacs/vscode etc). And what prevents spreadsheets becoming a platform? The install base of Excel, for a kickoff, but also the enormous limitations of csv on the other hand. In short, the conditions that allowed all those 1980s applications to flourish – bespoke file formats, invested users, no truly dominant players, less networked sharing - have all gone. And so have the apps.