The Myth of the Underachieving Genius

I don't think grades/school should be used for employment at all.* Grades may coincide with good skills but they are _COMPLETELY_ orthogonal.

1. Most programmers in the US go through 16 years of school. The final 8 have an effect on where you end up. Of the first four(high school) only about 1/3 of your time is spent on math/science* * . Of the final 4(at the two colleges I attended) only about 2.5 years are spent on math science. Of that 2.5 years only about 1.5 are computer-science specific.

This means you're determining somebody's programming competence based on information (What college and grades) that is less than half(48%) based on math/science proficiency and less than quarter(19%) based on computer science.

2. They're subjective. Especially liberal arts(i.e. MOST of school).* * *

3.It's incredibly expensive. Computer science and math cost next to nothing, but for some reason I'm supposed to give some school 60k over 4 years to spend about a quarter of my time studying what I'm interested in.

4.If you test out of a class it doesn't count for grades. If I work very hard and do 16 weeks of calculus in 3 and take a test, I get credit. If do nothing and just ace the class I get an A to pad my transcript. This creates a DIRECT incentive to do less.* * * *

* I am absolutely all for hearing about research/coding/contests I don't think schools opinions should matter at all.

* * I mention high school grades because high school grades determine which school you go to.

* * * Yes, math and science in high school/college can be and frequently are subjective as well. Teachers give credit for effort/showing work on problems/homework/etc. etc.. My high school algebra grade was a C despite getting 90+% right on exams because I didn't do the homework and the teacher wasn't satisfied with my explanations.

>As far as I can tell, very few people are arguing that people with lower grades are better than those with good grades.

http://news.ycombinator.com/item?id=1028272

There's some one arguing just that. While I don't see it a lot I have certainly seen the point argued, generally right beside some one claiming Einstein was bad at math.

I would agree that on their own grades are poor filter.

You could have less than stellar grades, but may have something else to show your ability (side projects or what not).

On the other hand, a high GPA might only demonstrate your mastery of bulimic studying; Memorize and regurgitate.

I don't know. My brother and me are a nice case study, of course that's just one more anecdote and not exactly science but still. People said that I would go to university and do all kinds of stuff, nobody paid much attention to Maurice. But he was the one to end up going to university, I dropped out in high school.

Family circumstances at the time had a lot to do with all this, but still, he's the one with the degree in history and I only have a typing diploma and a driving license as my 'educational record'.

Better grades or lower grades doesn't enter in to it.

The educational system works well if you're in the middle of the pack, if you are at one or the other extreme you have a problem.

I'm not sure if a university degree would have helped me in any way, I'll never know and I'm resigned to that.

But I do know that there is more to school than 'subject knowledge', and that in my dealings with people I could have benefited from a deeper immersion in to social structures and how to deal with all kinds of situations.

See my post the other day about my past mistakes, I'm pretty sure a few of those are rooted in my checkered history. I wouldn't trade my life for any other, and in the hiring policies of my company I always tried to look past the grades and to the person behind them and their motivation.

If I got the feeling they wanted to move ahead in life then that was usually the deciding factor, and for the most part they did not disappoint me, even the ones with good grades ;)

I agree with you on everything except the last bit. I believe the literature on what a good programmer is has been fairly clear - Fred Brooks first wrote about it in the Mythical Man-Month, and since then there has been a large body of empirical evidence supporting his claims on the variance between good and bad programmers.

Good programmers are simply more productive. They ship more code, given the same amount of time. That's a simple way of looking at things, for sure (you may argue about readability and maintainability and so on so forth), but when the difference is 20 times the average programmer, people begin to sit up and take notice.