I'd still disagree. Empirically, it seems that career success in programming happens to people who approach a technical problem in a novel way (eg - ZFS, gmail

I'd still disagree. Empirically, it seems that career success in programming happens to people who approach a technical problem in a novel way (eg - ZFS, gmail, Asterisk, off the top of my head). Communication is important, but it is secondary. Of course, if your definition of career success is ascending through various ranks of management, I can see your point.