Author and Pentagon advisor John Arquilla believes that today's big-weapon systems are wrong for modern battle and for future battle
Some of the main points:
how many British submarines did it take to pen up the entire Argentine navy? Two. The Exocet missile proved the slow-moving capital ship's vulnerability. Today, the Chinese aren't developing aircraft carrier battle groups, but brilliant sea-going mines that know how to maneuver, supersonic anti-ship missiles -- which means the Falklands War on steroids -- and super-cavitation torpedoes, which create a bubble of air in front of the torpedo, letting them move at hundreds of knots per hour. The Chinese have an explicit "swarming" doctrine that can best be characterized as sea power without a navy.
You can create space defenses that don't require offensive capabilities. ANGELS -- Autonomous Nanosatellite Guardian for Evaluating Local Space is under development. They're autonomous nano-satellites. ANGELS will allow us to move satellites to safer locations and be able rapid ability to reconstitute space assets.
Insurgents and terrorists: understanding the enemy as a system and trying to pull that system apart. Deny them the web would slow them down a lot. Data-mine Net exchanges within Iraq. For more security we will have to give up some privacy.
I think he is correct that the US military budget should be more efficient. Bigger air craft carriers are not needed. $100 billion or more of spending on large systems should be re-purposed.
Also, the chinese capabilities would cause problems for the US navy and armed forces but new US beam weapons, rail guns and sonic defences means that I think the US would still win. But there is no need to fight China. So why bust the bank trying to do that? Neither side would decisively win. Nuclear deterrent.