I am a huge fan of the idea of OpenDoc and component-based software in general, to the point that one of my side projects is writing a desktop environment in Co

I am a huge fan of the idea of OpenDoc and component-based software in general, to the point that one of my side projects is writing a desktop environment in Common Lisp that is inspired by OpenDoc is not a clone; the emphasis is not on document embedding, but on increasing programmability and flexibility while maintaining a GUI environment. It would be kind of like a modern Lisp machine running on Linux. I was also inspired by a Hacker News post (https://news.ycombinator.com/item?id=13573373 https://news.ycombinator.com/item?id=13573373) about the Linux desktop could have benefitted from component-based software, where some of the commercial problems of OpenDoc would have been sidestepped by the wide availability of free, open source components instead of building FOSS analogues of large proprietary software packages. But with that being said, I agree with you that OpenDoc would have been a tough sell in the proprietary software world for the reasons that you gave. There's also the fact that 1997-era Apple was in no position to challenge Microsoft head-on by promoting component software. In order for the Mac to survive, Apple needed the backing of large software developers like Microsoft and Adobe who were willing to continue writing Mac versions of their popular, industry-standard software packages. OpenDoc might have made business sense in the late 1980s or even in the early 1990s as an alternative to the growing Windows ecosystem, but by 1996 it was too late; Microsoft Windows was the clear winner in the marketplace by a long shot, and Apple was spiraling down the drain before Steve Jobs returned. That's why it was a very big deal at MacWorld Boston 1997 when Steve Jobs declared that "the era of setting this up as a competition between Apple and Microsoft is over."