Well maybe if the guy's idea was explored and iterated as long as the other, entrenched one, it would yield far better results. After all, the proof-of-concept often performa worse in the beginning, but so many awesome technologies started off that way.