My major where they predicted four jobs for every grad turned out to be four grads for every job.
So by contrarian thinking architecture is probably the best degree you could pursue.
In the victorian times, the technology booming was railways. There was always a market for railways (until the car came along, but that wasn't really for another seventy years), but the real issue was the number which they built. There was passengers, but not enough to support the frantic investment that people were making (it was, after all, the next big thing!). This is where I see the VC backed SaaS startup investment market going. Frankly, there cannot really be a market for every SaaS app we see traipse through here with $10m of backing. The real question is, how long until the failure rate becomes high enough that people panic?
YCombinator has the right idea. Start with a small seed, build a good app and perhaps only once you have something launched, start looking for more investment. A website can generally lead to profitability either by being loss making but really, really popular, or by having revenue from day one. What worries me is the websites that have very little revenue from the beginning, and yet are attracting multi million dollar investments. That cannot last.
As an aside, we're thankful for those rail investments now - the UK not only has a very large railway network, we also have a very large bicycle path network built on ex-railway lines!
http://www.daapspace.daap.uc.edu/~stedmabn/profprac-base/hot...
The 2008 synopsis is here: http://www.areforum.org/forums/attachment.php?attachmentid=5...
Keep in mind that this was during the boom, and the average department head made less than $100k a year - that would typically be someone with at least 10-15 years of specialized experience and willing to take on personal liability for architectural designs under their seal until the statute of repose (varies by state from seven years to forever).
Journalism is the 'hot' major they encouraged. Personally I think college career advice is similar to relying on 14 day weather forecasts here in Michigan.
I had no interest in computer science because I had no desire to wear a white coat in a frigid data center and find hanging chads in punchcards. Yet a few years after graduating I bought one of the first IBM PC's and was using Compuserve and coding in dBase within a year.
Most people graduating from college today are going to have four separate careers during their working life. You need to keep the habit of lifetime learning because the world will change in many interesting ways.