Why CS people should go to college - lessons from basketball

> One is that doing this is very rare but secondly there were many more who jumped in early who disappeared quickly either into limited roles or out of the NBA and no one remembered them. What he pointed out was that playing college ball was an important part of learning the game as well as growing up. Players who skipped college often had only their natural talent which while it might be significant limited them at the professional level.

As an NBA fan... that's not really true. If you look at the list of players who skipped college, it's filled with superstars like LeBron James, Kevin Garnett, Kobe Bryant, etc. Even the less spectacular ones still all had productive careers with high paying contracts, like CJ Miles and Lou Williams. Abject failures are very few and far between. Gerald Green is the only one that immediately comes to mind.

The only reason players don't skip college now is because the NBA doesn't let them anymore. They set a minimum age limit.

So there's two things here, the notion that developers need to go to college, and the analogy to basketball via the notion that NBA players are better off after having played in college. I'm going to expand and examine that analogy first and then draw it back a bit.

The reason for college basketball (as a talent pipeline to the NBA at least) is because college basketball incurs no cost on the NBA or its teams to operate, just like college football for the NFL, but unlike minor league baseball or soccer academies in Europe and (secondarily) run by some MLS teams. The average professional athlete (by which I mean someone who makes their sport the focus of their lives, rather than someone who gets paid), in most team sports, is only ready for top level play in their early 20's anyway, which roughly corresponds to college graduation. So it works out.

I don't know much about minor league baseball, but I do know a bit about soccer. In soccer, different players develop at different rates. Some players, like Lionel Messi or Wayne Rooney, can start as young as 16. More often, a player will break into top-level play between 18 and 21. While some of the more special and talented players will start younger, most players, some world-class, broke through after age 20; goalkeeper Joe Hart broke into top-level play at 22 in 2009, and defensive midfielder Sergio Busquets broke through at age 20 the year before. As far as anyone could tell, this is because those are the ages that those players were ready. But this is because youth players come through academies run by the clubs themselves. If a player is ready at 16, they'll play him. If a player is ready at 22, they'll play him. Just like sometimes a basketball player is ready at 18 and doesn't need college ball.

Of course, it's worth the expense to develop players in specialized academies when players can be worth millions; developers aren't so valuable. But they are just as individuated, and some will be ready at a different age than others, and college isn't necessarily the best way to prepare them.

The difference between college and the professional world in CS versus basketball are quite different:

> In college basketball, you play basketball. In pro basketball, you play basketball.

> In college CS, you write theoretical solutions to theoretical problems. In the pro world you write real solutions to real problems.

That aside the argument and article are weak at best. I'm having trouble understanding what exactly we're supposed to take away from this aside from "College good! Skipping college bad! In basketball, and therefore CS!".

There's not a whole lot of substance in this article. It's a bunch of opinions that are unsupported by any evidence.

It may be true that college confer benefit, but that needs to be demonstrated.

The article mentions 'learning to learn' amongst the benefits of going to college, but surely a programming 'superstar' who didn't go to college would know a hell of a lot more about learning on their own than someone who had a tutor all throughout their college years.

I didn't finish college specifically because I knew how to learn on my own. I acquired the skills necessary for my job on my own time.

Sure, whatever.

You still need your 10,000 hours of practice to be a god among men. That's 10 years at 20 hours a week. This means you need to start programming at age 8-12 if you want to be a superstar early.

It's what the basketball stars are doing, right?

And by the time you've been programming 20 hours a week for 6+ years, college isn't gonna teach you a whole lot.