This idea of intermittent rim braking has been around since Cycling Day 1, and I say it is a myth. Here's why:
In a nutshell: no matter how you brake, you are dumping the same amount of heat into your rims and they will get just as hot one way or another. The reasoning is certainly not original with me:
Physics 101: To get your weight from up here to down there, you need to get rid of a certain amount of energy, fixed and determined by the weight and the vertical descent. Wind resistance dissipates some of it, heating the air you pass through (and your body a little bit that wind chill overcomes). Rolling resistance converts a bit more into heating your tires. The rest goes into the brakes, heating the rims and the brake blocks.
Your style of braking does not change the energy dissipated by wind resistance and rolling resistance. Only your speed affects those.
Consider, then, two bikes of equal total weight descending side by side. One rider brakes steadily and one brakes intermittently. They both need to lose the same total energy. Because they go the same speed, they lose the same amounts to wind and rolling resistance. By subtraction, they also lose the same amounts to their rims and brakes. And because they go the same speed, they have the same time in which to absorb or dissipate all this energy as heat.
But we do not care much about energy or heat. We want to know how hot the rims get. Given a fixed amount of heat, what affects the temperature rise?
Certainly the heat capacity of the rims and brake blocks. A big, heavy rim can take more heat before its temperature rises--say 50 degrees--than a lightweight rim, the amount being roughly proportional to the amount of metal to be heated, that is, its weight.
Contact between brake block and rim assures that the blocks never get hotter than the rim. The blocks, being much lighter than the rims and made of rubber rather than aluminum, have such a tiny heat capacity that they absorb very little of the heat.
A black rim radiates more heat than a shiny rim. If they were red hot, the difference would be significant, but not much at all at our temperatures.
One might argue that letting some air between brake block and rim helps to cool the block. True, but note that neither rim nor block is absorbing energy then. When the block hits the rim again, it reaches rim temperature in milliseconds. Now they have less total braking time to absorb the same amount of heat; the temperature just has to rise faster.
Sheldon Brown writes, "On long, straight mountain descents, ... pumping the brakes, alternating between one and the other, will briefly heat the surface of each rim more and dissipate more heat before it spreads inwards to the tires." (http://www.sheldonbrown.com/brakturn.html
) Also true, but a tiny effect. How long does it take heat to pass through the 1/16" of aluminum between brake block and tire bead? Less than a second, I'd think.
The two ways to really keep your rims cool: 1) Increase the wind resistance by sitting up, spreading your elbows, and descending faster. 2) Provide more time for the rims to lose their heat by radiation and convection, that is, descend slower.
Most of us cannot descend the hairy grades fast enough for option 1 to work. (Racers do it, though.) We hate to creep down the mountain, so we descend at a moderate pace, stop and wait for rims to cool, and get to the bottom at about the same time we would have made while creeping.