When I hear a high-tech entrepreneur tell me, "If I can just raise $250K to build a prototype..." I know he or she is never going to make it.
You used to be able to raise venture money based on an idea and a team, but those days are long gone unless your name is Ev Williams or something similar.
It's not that VCs can't make money by investing in ideas; it's that it's so much more efficient to invest in companies with traction. Don't waste your time (and that of us angels and VCs) trying to raise money before you get traction.
(And no, this doesn't contradict my remarks from earlier today on Vivek's "VCs Harm Innovation" post; there is a lot of risk remaining in a startup, even when it has traction. Both PBworks and Ustream (the two most prominent companies I've been directly involved with) had a ton of work to do after gaining initial traction. For both companies, the initial traction enabled raising seed money to get them to a traditional VC investment.)
As for the non-technical folks who wonder how they're supposed to start a company, I feel your pain. I can't code, and that's a major disadvantage. But great code, while necessary, is not sufficient for success. Building a better mousetrap is not enough; even getting TechCrunched is insufficient (though helpful!).
To overcome your impractical choice of college major (before you flame me, I freely admit that I have a degree in Creative Writing...not at the top of most VC wishlists) requires work, but isn't impossible. Get to know brilliant coders, build long-term relationships, and demonstrate that you can add value.
EtherPad (blatant plug--another investment!) recently hired Daniel Clemens as COO. Why? Because the founders had worked with him before and trusted him.
And if you can't code, and you can't get brilliant coders to like and respect you, you probably shouldn't be a high-tech entrepreneur. There's no shame in being aware of your own abilities and steering your career accordingly.
P.S. Bear in mind that this advice only applies to high-tech startups--there are plenty of great businesses where a CS degree is not an asset!
(Inspired by a post Seth Sternberg)