Your claim is as hard to believe as his. With a master's degree in C.S. I started at about $65k in 1999. At IBM. In a market with a much lower cost-of-living than California and with correspondingly lower salaries. And the position I was hired into was pretty junior. So it's hard for me to believe that a guy with a Ph.D. getting a job in a hot field and in a hot market would come in making $85-90k.

The kind of dev. I'm interested in doing isn't being outsourced. The type of jobs being outsourced aren't the ones I had in mind when I chose to pursue a C.S. degree.

I got credit for mine. Let me skip the intro course. Even if only gets you credit for a course intended for non-majors that can still potentially boost your GPA. Having the high score might also help you get admitted to more selective schools. Not taking the test when you have a reasonable chance of scoring highly seems like pretty terrible advice.

My experience isn't comprehensive, but I haven't personally heard about any shops that still make extensive use of perl for new work. I'm not calling you a liar; I guess they're out there.

What we'd want to see is a ranking of languages by "new project starts" utilizing that language. There's still COBOL around but how many new projects are started that use COBOL? Etc. I suspect few people starting a project today that requires a Perl-like language would actually choose Perl unless they were already a Perl expert and it was definitely going to be a one-man job. They'd choose Python/Ruby/PHP instead. So, in that sense, Perl is dead.

"Genes don't just predict height, but also how good you are at basketball".

As I understand the summary, the researchers explained 20% of the heritability of height. Not height in general. If I haven't misread things, then 80% of the portion of height that is genetically determined is still unexplained. Diet no doubt has a huge effect on height, but (again, if I'm not misreading things) that's not what's being discussed here.

As a lay person, that a study involving 250,000 participants could explain only 20% of height heritability seems like a bad sign with respect to the pace at which we're likely unravel our own genetic code.

Maybe. Some bright/bored people are nevertheless not willing to let down team members. I count myself in that category. In high school you typically don't have team members, so you're not letting anyone down (except maybe your parents) when you opt out of homework, problem sets, etc.

We could have required only pseudocode on the exams, but if so then we would have had to "standardize" on a particular grammar of pseudocode or it would be near impossible to grade students' answers. You'd have students arguing about what their own peculiar version of pseudocode "means". So, since all the programming assignments were required to be done in Pascal, we just adopted "Pascal" as our pseudocode for the exams. We weren't picky about syntax unless a question was designed to test syntax. If you forgot a semi-colon or curly brace you didn't get points deducted. Also, this requirement (Pascal) was communicated clearly prior to the exam. Engineering guy just figured it didn't apply to him, I guess.

If the set of all students consisted only of Steve Jobs then you would have an excellent point.

I actually think essays have the opposite effect. The presence of an essay on an application gives an edge to the student who can bring the most resources to bear in "crafting" that essay. Perhaps, at the extreme, having it ghost-written by a professional college admissions consultant. I'd almost rather it boil down to class rank and test scores. Possibly with the addition of a space for students to list relevant accomplishments in a non-essay format. "First-chair trumpet at state", "state science fair winner", etc. Teacher recommendations are a little suspect as well, since schools (and therefore teachers) have a vested interest in seeing their students admitted to prestigious universities. There may also be students who are qualified from an academic perspective but have failed to sufficiently ingratiate themselves with their teachers. Or maybe their teachers, as a group, don't know how to write "compelling" recommendations (possibly because one of their students applying to an elite university is a once-in-a-blue-moon occurrence).

Also, when it comes to test scores, I'd prefer to see colleges use something more content-related instead of the SAT, which is pretty easily gamed. Maybe something similar to the A-levels in the U.K. or the AP exams in the U.S. If the AP exams were used there would need to be a new math exam created that stops short of Calculus; it's not reasonable to expect every college student to have taken calculus. AP tests are, arguably, harder to "game" than the SAT. In theory you have to actually understand the material.

Not that I necessarily agree with it, but I think the explanation for "bright C students" is that they're creative geniuses who are so bored by the "standard" curriculum they just decided to "opt out". Hence the low marks.

Please overhaul! But not out of fear the next Lucas, Spielberg or Jobs isn't going to be admitted. Do it because it's an annoying waste of time and effort to fill out a completely different application and write a completely different set of essays for each and every school. Even better, establish a single application fee that buys the student the ability to apply to some (reasonable) fixed number of schools. Believe it or not, the cost of application (esp. when applying to several schools) is actually a meaningful disincentive for students at the low end of the income spectrum.

You sound like a guy I had in my discussion section when I was TA'ing an intro Computer Science course. (Taught in Pascal). There were a handful of questions on one of our exams that asked the test-taker to write some code. He wrote his in C, because he never actually bothered to learn Pascal syntax. We gave him zero points. Guy was a third or fourth year E.E. student who was just taking the intro C.S .course to pad his GPA.

Point being: your acquaintance's teacher would probably (can't say for sure) welcome him or her exploring new aspects of the language on his/her own. However, when working on exercises designed to highlight a particular language concept (possibly a fairly simple one) it's not acceptable to pull in more advanced concepts that completely obviate the concept being taught.

For instance, if a question asks you to use a for loop to do something N times and you instead call some library method that does the same thing without using a for loop (i.e. probably the better solution in a real-world scenario) you haven't answered the question correctly. It is designed to test whether you know how to write a for loop, not whether you know how to solve that trivial problem in a novel way.