2008 April

When the sun is high, the flowers bloom, and the baby cows roam wide, a young cook wakes compelled to make elixir. The Ancient Rite of Veal Stock begins:

Restaurant Depot makes this easy. 30 lbs of “Veal Mixted Bones” coming up…

My pot is not ready for this.

“Skim, skim, skim!” Keller says.

The now-poached bones, drained of yuckiness

Everyone’s favorite aromatics. And an obscene quantity of thyme.

Back in the pot, everyone! Tomatoes, come join us! The water’s fine!

We need smell-o-vision so that you can know how happy my apartment was with this.

Reduced!

Omitted from the photoset: several phases of straining the veal stock whilst straining with heavy pots.

Not quite as much yield as I would have liked, but I still have 15 pounds of bones. Next time: a roasted stock!

I’ll tell you a secret: Caltech’s admissions department will tell you that the school has superior academics and will provide the best possible education. That’s really not true at all. I came here because everyone is so insanely good at what they do.

I’ve been thinking about the source of their talent for a while now, and I think it has far less to do with their education than any college cares to admit. Let me tell you a few stories.

My Roommate, the Math Major

A lot of people say they have crazy roommates. They’re wrong—I have a crazy roommate. Not in the sense that he’s unstable or anything. We don’t have passive aggressive post-it fights. Mike’s just truly and frighteningly smart.

At the beginning of the year, I was playing Lights Off and struggling to pass an early level. Mike came over and casually watched me flail about. Suddenly: “Huh. Can I play with that?” After a few taps, he continued: “Oh! It’s degenerate.” He walked over to our whiteboard, scrawled some incomprehensible runes, and then proceeded through the remaining two hundred levels in about three seconds a level.

Mike started to explain the generalized solution, but I was glad I didn’t follow—this way, I can still enjoy the game!

Alex, the Pianist

Until he went off to join the Korean army a few weeks ago, my a cappella group had been graced by the presence of a pitch-perfect jazz pianist named Alex.

Sometimes in rehearsal, he would hold impromptu karaoke sessions, playing grandiose piano accompaniments to any song we could name. Everyone else would sing along and have a great time, but my gaze would never stray from his hands.

Alex added embellishments to every song, making each more musical and impressive than their crude pop songwriters could imagine. Extra 9s, 11s, suspensions, and harmonies danced freely throughout.

Every time he played, I asked Alex how he did it. Well, ask a stupid question, get a stupid answer:

“I have no idea.”

A Programming Tutor

I’ve helped—or at least tried to help—quite a few people write code over the years.

Programming can be a tricky thing because there’s a lot of concepts that need to be learned simultaneously to get anything done: how to “speak” a language, system concepts, and most importantly, algorithms.

By algorithms, I don’t mean quicksort and big-O notation and the like. Just the notion of translating a problem into something a machine can solve.

A common introductory problem for students learning about arrays is to produce the average of a given list. Intuitively, the solution is easy: sum all the elements and divide by the length. Yet I’ve seen so many people suffer with such a simple task, and not because they’re getting tripped up by language syntax. They all miss the critical insight of using a temporary “working sum” variable.

It’s not a matter of being smart. It’s not a matter of missing knowledge, per-se. When I consider this algorithm, I produce the solution without thinking about things like working variables or enumeration steps: I’m just translating between the language in my head and a language the machine can understand.

Why and How

Mike didn’t look through any books to find an answer to Lights Off, Alex didn’t read any sheet music to play his accompaniments, and I didn’t look up sample code to implement my algorithms. We just knew, like anyone who’s good with his trade just knows.

That’s what I’ve noticed during my time here at Caltech: real skill comes from intuition.

And the intuition comes from, it seems, a mixture of experience and innate predilection. Mike’s been doing math since his childhood: he carries around paper and a pencil at all time to work on problems in his head. Alex and I have similar stories about what we do.

Now, the interesting thing is that I’m not entirely sure where education comes into this. I’m sure it’s important, but I think it’s really just a tool, rather than something that actually creates skill. It gives us techniques we can apply to get more experience, but those techniques are useless until we’ve built up intuition.

Sure, a good professor will give his students those techniques a little faster, but I’m convinced a book could eventually lead to the same skills as a six-figure education.

Caltech gives its students the chance to be around others at their level, but I think that’s more fun than educational. My fellow students haven’t helped me be a better software engineer—they just keep me sane.