I am not sure you are right about young developers versus 'old' enterprise developers. (for the sake of argument, i am not thinking that way) To actually succe

I am not sure you are right about young developers versus 'old' enterprise developers. (for the sake of argument, i am not thinking that way) To actually success replacing SAP, for instance, you must support legacy applications, old modules, and various minor modifications for tiny little features demanded 10 years ago which nobody really knows why needed in the first place. Startups tend to go for the 'easy'/'cool', 'brilliant' new idea and not for the boring, tedious huge applications, simply because it's faster to build new ideas. Take a look at the open source operating system community for instance. There is no real big 'new' development there, and neither in the compilers, debuggers and C runtime libraries. No one wants to rewrite it. It simply work. There are two major compilers (correct me if I am wrong) in the open source community and no one writes a full scale compiler simply because he has new idea about how to generate code faster. It is somewhat the same in the enterprise area. People like to do easy work, that's all.