Climbing is a different matter though - aerodynamcis play a smaller part in the equation and weight and flex do matter.
Weight matters, but flex only matters if you believe that the frame is a damped
spring. This is more likely to be true of a carbon frame than a steel one. See http://www.bikethink.com/Frameflex.htm
In fact, to the extent flex exists it might actually *help* deliver power:http://www.kirkframeworks.com/Flex.htm
Remember that carbon frames (or high end ones) are designed to be compliant in certain planes to give a more comfortable ride - it's the difference between the inherent characteristics of steel and a frame system specifically designed to give the bum an easier ride.
People over rate the inherent properties of material in frame design, although usually its the steel-forever crowd who do so. Modern carbon frames certainly can be comfortable, but isn't it comfort that keeps the boutique steel frame makers in business too?
Flex does have an effect on cornering and steering at speed. Quite a noticeable one.
That is true!
Finally - longetivity - steel frames for racing used to be finished after a season. That's why the teams gave them away. That's not true for carbon - though in top end use they'll only have probably two seasons use. A racing frame is not built with longetivity in mind.
Yes, racing frames aren't made for longevity, whatever material they are made of.
Back in the real world, however, steel is the longevity king. Why? Because it has a "soft" failure mode - i.e. it fails gradually, by bending. Alu and carbon otoh are prone to go "snap!" This matters a lot more with a secondhand bike, of course, because you don't know how it has been used.
I'm not anti-carbon by any means: I think it makes a lot of sense in a high end racer where it's worth paying big bucks for marginal performance increases. But some of the claims made to get club racers and commuters to pay extra to ride it are pure bunkum! (Excuse my language.)