Windows as a product feels that way, but I think if you're a kernel hacker, that's not really true for you. Monolithic kernels for Unix-like operating systems like GNU/Linux aren't fundamentally that innovative either. (There's innovation within Linux, of course.)
I also don't really think computing advances in such a linear way. Lots of cool new tech is about digging up underappreciated insights from computing's distant past and applying it in a new context, or even just propagating it more widely.
I'm not saying Windows 9x in particular had anything super interesting going on. But all of the viable desktop and server operating systems are based on really old tech, and at the same time computing's distant past is full of hidden treasures.
I also don't really think computing advances in such a linear way. Lots of cool new tech is about digging up underappreciated insights from computing's distant past and applying it in a new context, or even just propagating it more widely.
I'm not saying Windows 9x in particular had anything super interesting going on. But all of the viable desktop and server operating systems are based on really old tech, and at the same time computing's distant past is full of hidden treasures.