Great post by Euthymia.
I'd already been in the biz for 10 years when I first saw Mosaic, but I dismissed it at the time, thinking it was just a BBS for eggheads. I'd similarly dismissed Windows 2.0, because it couldn't run existing software. I remember seeing my first Mac, which a customer was excitedly showing off, and thinking "grade schools will love this, but it'll never have a place on corporate desks". When the first smartphones came along, they struck me as gimmicky toys trying to do many things, but none of them well ("no way that'll replace my digital camera").
When I started using RISC-based and massively parallel and distributed systems, with their potential for cool stuff such as real-time animation, I thought "now,
there's the future!". My Sun Sparcstation was way ahead of anything Intel, Apple or Microsoft were up to at the time. I was looking forward to everyone having something like that in their homes. You could even record and edit high-fidelity audio on it.
My point is that my own informed prescience doesn't have a great track record. It's just not always possible to evaluate new approaches within the framework of the current paradigm. Too many assumptions about what works and what doesn't, based on what's working today.
Heck, I thought Rap and Disco were passing fads. I was, however, spot on when I predicted the demise of parachute pants and leg warmers.