Would not the top of the pyramid intersect with the bottom in the land of computer science research, where new ideas are being developed but the software itself

Would not the top of the pyramid intersect with the bottom in the land of computer science research, where new ideas are being developed but the software itself is often hacked together to the point of just barely working? Not sure if this is a useful observation, or only an observation.