Agree with all of above, all basic physics and ski prep. But less clear why straightlining is criterion. Nor why a question about construction ends up with base prep. Seems as if that should be held constant.
If OP tries the longest radius turns he can produce, nearly a straight line but on the edges, he'll not only be moving a lot faster (and more securely), but he'll be able to bring more of the ski's construction into play (flex pattern, energy storage, vibrational resonances in an arc), not to mention his leg strength and skill set countering basic mass of his buddies. I'd guess that stiffer skis "go faster" not only because they can allow higher speeds before they lose contact with the snow, but also because they can store more energy for release at the end of the turn, and (probably) damp down oscillations that are wasted energy. Stiffness=construction.
An allied question is whether designers optimize for straight lines or turns. I'd guess turns, given the attention to flex patterns, shape, and so on. The perfect straightline ski would be, ah, straight. And not flex at all. Lot like the planks I skied on in the 50's.
Finally, unless I missed it, no one's mentioned length. Not precisely construction, but linked to construction. Flex patterns of GS and up racing skis, for instance, are specifically designed to optimize longer lengths than rec skis. Which in turn tend to come in longer lengths/greater stiffnesses the higher performance they are. If there's any basis for analogy with boat hulls, perfectly stiff longer skis in anything except glare ice will simply have a higher possible terminal speed in a fluid medium. (Yeah, I know there's a debate about whether snow is a fluid or a bunch of particles, but we still use measures of behavior in fluids, like float, to talk about skiing. Would like to know from an engineer who knows hydrodynamics if the analogy holds.) And obviously, longer skis will be more stable.
Edited by beyond - 7/14/10 at 4:56pm