Trying to please everyone rarely works
The problem is there are two camps that use OpenGL - both with fundamentally
different needs.
The game developers want access to the latest hardware technology and don't mind re-vamping their rendering engine often to get it. The graphics are often what sells a game and so to get that 'wow' factor game developers need access to the latest feaures asap - they want a stripped down 'lean and mean' OpenGL so that GPU drivers can support new features easier and more quickly. The fact that in 5 years time these current features will be obsolete and newer drivers will have dropped support for them - meaning the game will no longer work - does not bother them.
The CAD developers 1st and foremost want a stable platform that just works, and will continue to be working in 5 years time - the life of a CAD project can be much longer than that, so the original software used *must* still work (often unpatched) even on new hardware with new drivers. Speed is important, but not the main objective - CAD developers spend most of their time developing the user interface and making the CAD functionality better / correct. They want a graphics API that is extensible but rarely loses core functionality.
I'm sure Microsoft and others opposed to the cross platform and 'open' nature of OpenGL are laughing their socks off at the current situation of trying to please everyone and ending up pleasing no-one.
As I see it, the only way both camps will be happy is if OpenGL is forked into two separate libraries - a fast evolving 'gaming' library and a more sedantary 'CAD' library, both of which could be kept small enough for both GPU driver writers and hardware to support, fully accellerated. Most GPU manufacturers already have separate product lines (and often drivers) for gaming / CAD apps so there really shouldn't be any problems having separate OpenGL's.