I note that in it, Stephenson makes a prediction, as if it were a known fact, that I think we now have significant reason to doubt.
He says "Rockets are as close to perfect as they're ever going to get. For a few more billion dollars we might be able to achieve a microscopic improvement in efficiency or reliability, but to make any game-changing improvements is not merely expensive; it's a physical impossibility."
I think we now have reason to believe that for around a billion dollars ($1b is the estimated total R&D budget as of a year ago), SpaceX has already made better-than-microscopic improvements in efficiency. But more interestingly, there's reason to believe that with little or no additional outside investment (i.e. working on profits), SpaceX is pretty likely to soon make a very large improvement in efficiency, with Grasshopper &c. Rockets are probably not as perfect as they're ever going to get.
Still a very impressive achievement, but looking at the performance of SpaceX's engines circa 2012 and comparing them to what the Soviets built in 1960 or 1970 helps put the pace of the industry into context.
The relevant efficiency, to me, is "awesomeness achieved per human effort and non-renewable resource", and the useful proxy for this is "awesomeness per price". Arguably SpaceX rockets are cheaper (per given mass to given orbit, obviously) that most or arguably all of the competition. I say "arguably cheaper" because it's a little tough to figure out how to price the shoulders of giants on which they stand.
And really I only picked the word "efficiency" in the context of Stephenson's "efficiency or reliability".
Anyway, hopefully this explains why I think that true reusable rockets, Musk's often-stated ambition, qualify as a "very large improvement" in efficiency. Specific impulse is not expected to significantly improve, right?
Also, I apologize in advance for my legion failings in the area of comprehending aerospace engineering.
[1]: http://www.slate.com/articles/technology/future_tense/2011/0...