I agree with you here. This is a contrived example - I would expect a bit more from an "article" voted to the front page of hacker news.
Sure, you shouldn't hire a developer merely because he went to the right schools and did well, but I can say with some certainty that if you compare candidates who went to good schools and got good grades with candidates who went to lesser schools and got poor grades, the former will as a whole be much stronger.
While you will hire some bad candidates and skip some good candidates if you work this way, your strategy will be pretty close to optimal. Some very successful companies have been using this strategy for many years.
With that said, if this company thinks they have a better way to do things, they are welcome to give it a try. If they can improve hiring even a little bit (either in improving the positive or negative filter), I'm sure there is money in it for them.
Nope. I know developers who don't even have degrees that you'd be /crazy/ to ignore.
On the average you will do pretty well with your "good school, good grades" strategy. However, I believe you will miss out on the occasional truly extraordinary candidate, the one that was bored with school and quit to do Neat Stuff.
But (you may argue) these people are prima-donnas, or easily-distracted flakes, or won't fit in culturally, or they don't bathe. Or something.
Look at their track record /outside of school/ and see if they did indeed do anything Really Neat. If not, they were fooling themselves. But if there's evidence to the contrary, you're crazy to ignore it.
Frankly, I don't even look at the education history any more. It doesn't tell me anything. I'm far more interested in what projects they've done and their exposure to computer science and software engineering than what dorm they had an opportunity to occupy.
Sure, you shouldn't hire a developer merely because he went to the right schools and did well, but I can say with some certainty that if you compare candidates who went to good schools and got good grades with candidates who went to lesser schools and got poor grades, the former will as a whole be much stronger.
While you will hire some bad candidates and skip some good candidates if you work this way, your strategy will be pretty close to optimal. Some very successful companies have been using this strategy for many years.
With that said, if this company thinks they have a better way to do things, they are welcome to give it a try. If they can improve hiring even a little bit (either in improving the positive or negative filter), I'm sure there is money in it for them.