Most applications that are being developed are 100% standard fare (CRUD).
Also a lot of programmers seem to get by just fine without a CS degree. (I've got a maths degree and minored in CS, but still, sometimes I feel I wasted my time). In fact I don't think I have ever been asked by an employer to code something complicated. If I were doing a startup it might be a different matter.
Are you saying that you shouldn't bother to read the PickAxe book or the PragProg Rails guide unless you have a CS degree?
Also, most code is probably crap but works anyway. I suspect the same can't be said for appendix removals.
If you have said something like "developing CRUD web apps is so easy compared to some other jobs", I would have had nothing to say to you, because you're probably right.
Maybe if you had gone to work at IBM on Watson, or for a NASA research center on spaceflight communication networks, or work for something like Certicom producing the next public key cryptosystem over elliptic curves, you'd really be flexing your math muscle and programming. I can imaging there are a small subset of startups as well that would extremely value your math capabilities. So you feel your degree was wasted modulo the way your career panned out.
And a final note about code being crap: Many dental and medical tourists come back to the US with extremely poor workmanship on their knee replacements, fillings, or root canals, and have to have much of it re-done here. Also, due to the nature of medicine and software development, the comparison isn't really apples-to-apples. An oral surgeon might do 10 wisdom-tooth removals a week, granted everyone's mouth is different and there are a huge number of complications, but many software developers are working on something entirely novel on each project. In my experience, crap code is a function of changing requirements + iterated attempts at refactoring.
The basics of medicine (particularly of routine surgery) are not particularly hard to grasp; the hard part is learning to control the scalpel, and that's really not very different from learning how to cut a frisket mask for airbrushing. Sure, there are hard parts, but how many GPs ever have to delve into the deepest arcana? That's what specialists are for.
Programming is easy if you're already inclined to think that way, or once you've had the "aha!" moment. Getting there isn't universal, even among those who have chosen to pursue programming as a career. Pick a platform, visit a forum or newsgroup, and look for the "kindly do the needful" messages from people who are paid to do things that should be as simple as putting two Lego blocks together.
[1] http://en.wikipedia.org/wiki/Ferdinand_Waldo_Demara [2] http://thedailywtf.com
Maybe it is because both of my parents work in the medical profession, but that sounds perfectly sane to me. Most of diagnosing a medical condition is applying Bayes theorem to symptoms. Have symptoms x, y, z? It indicates condition a, or b. Run a few standard, well known tests that lead you to believe that is a instead of b. Follow standard treatment for a. Say a is gal stones, a few weeks of reading would make you close to an expert on gal stones, treatment, causes, how to diagnose etc. Now sure if a is brain cancer then it is a little more involved but there are corners of CS I would say the same thing about.
High school biology wasn't very hard, and I imagine you could be minimally useful after little more than that. But to actually become a doctor is a very different matter. Just like being a Java jockey is very different from working on high-performance distributed systems.
Given pre built systems, everything is simple. And since "software is eating the world", maybe everything can be simple.
I can't help thinking about medicine instead. Admittedly I haven't tried, but it seems unlikely that the same thing would work for medicine (studying books for a couple of weeks). Even learning medicine in isolation (as is common for programmers) seems like madness.
I will now go looking for tutorials on removing an appendix on YouTube, but I kind of hope that I won't find much...