My work involves building large, complex automation systems for Fortune 50 type companies. We have recently branched out to offer overhaul/rebuild of our competitors equipment (they are, fortunately for us, not interested in servicing their own stuff). We have a piece here circa 1997 at the moment. From a technology point of view, it was obsolete in the late 1970's. They obviously have held on to a design for far too long. Last year they sold no systems. We won every bid.

It's obvious to me that as a company, particularly a North American company, you innovate or die, as my competitor did. I've been mulling this idea over and trying to parallel it to IT in general and OS design in particular. What was the last neat or great thing that Linux has innovated? 2.6? The new CFS (scheduler)? Nothing? How about Windows? Vista?

If I ponder long enough I will think of some, I'm sure. The neatest thing I've played with in some time was OSX. All of OSX's flaws and flames aside, the concept of the mach microkernel with another kernel running atop seemed kinda cool. The execution in OSX seemed to leave a lot to desire (I referred to it as Frankenstein when nemo was giving me the guided tour during his research for the phrack article).

I would hate to see OS's like Linux slow innovation to the point where they become some sort of ancient Rube Goldberg contraption like the beast on my shop floor, but the new corporate control of the development (read the domains of the maintainers) may head it there.

If none of these folks innovate, will it be stagnant, or is it ripe for the next 'Linux' to come along and create the new innovation.

More thoughts later.