The big problem for OpenGL was (a) Windows platform dominance, particularly in gaming, and (b) weak support for rapidly improving 3D hardware (but that was larg

The big problem for OpenGL was (a) Windows platform dominance, particularly in gaming, and (b) weak support for rapidly improving 3D hardware (but that was largely a case of already in progress network effects -- OpenGL drivers suck because very few people care, so even fewer people use OpenGL). OpenGL was kept on life support in the aughts by high-end 3D products which are often cross-platform (e.g. Maya), and Apple (which switched to OpenGL from QuickDraw 3D after the spectacular failure of the latter -- incidentally OpenGL being "open" was SGI's proactive defense against QuickDraw 3D), and then rebuilt the entire Mac UI architecture on top of it. OpenGL ES and iPhone have completely revived OpenGL's fortunes -- there's nothing much to worry about now unless Android and iOS both fail in the marketplace. (It's entertaining that the Stackexchange post linked elsewhere was written more-or-less as a final "this is why OpenGL died" -- it's not dead yet... getting better...)