My First Weeks at Caltech/NASA, and, What Am I Doing?

This summer I landed a gig developing Scientific Visualizations for three projects at Caltech and JPL (The Jet Propulsion Laboratory, owned by NASA). I actually started back on June 18th but it took a while to get settled in and rolling in a way that resembles some sort of routine. So, I’m living in Pasadena, exploring the area and working with a team of crazy people.

First things are tautologically first: Transportation.

Last semester at Oberlin I was inspired by Aaron K. and Benjamin GH to learn how to rollerblade. They used inline skates to get around campus and it seemed like a super fun way to get from place to place – Aaron described it as ‘really enjoying the way you move’ – so for a birthday present I asked for a pair of Rollerblades, with the intention of using them to get around Pasadena.

Blades, Pads, Guitar, Dr. Zhivago.

Blades, Pads, Guitar, Dr. Zhivago.

Let me say, there are some things about rollerblades. The first is hills. When on rollerblades you develop a very good sense of which way is down because if you don’t then you start going there pretty quickly. That leads to the second thing, which is stopping. There are several methods for stopping on inline skates, with the most popular among beginners (e.g. me) being the heel brake. There are many other ways to stop, like the T-Stop, but these all involve some form of balance and coordination, otherwise your body becomes the brake. Combined with the ‘turning’ learning curve, this makes inline skates a very interesting thing to start using. That’s where the pads come in. I think the wrist guards are the most ingenious thing ever. They’re reinforced with thick plastic on either side so that when you fall you don’t skin your palms on the pavement. The plastic on the side of your palm is bumped out a bit so that it hits the ground first. I want to hug whomever came up with that design.

Wrist Guard

Best idea ever

The other thing about Rollerblades is crossing the street. Sometimes you just can’t keep going through an intersection because it’s currently being occupied by cars. In these cases the issue switches from stopping to getting going again. At first I stuck to the sidewalks, and in order to travess the street as a pedestrian I had devised the following elegant maneuver:

  1. If the full entourage  of pads and helmet have not already tipped off the locals, you must make it patently obvious that you have no idea what you’re doing. Experienced skaters are usually expected to cross quickly and get out of everyone’s way. So long as people can see you’re not one of those, you can invoke the “aw look, he’s trying to walk” response from motorists which may buy you some sympathy or at least some time. This technique has been leveraged by babies for all of humanity’s existence and we are all quite obviously alive now, so something about it has to work.
  2. When the signal changes, grab onto the traffic pole. Depending on your mood you can make it more theatrical or utilitarian as needed. Edge one skate against the pole to keep it from rolling away while your attention is elsewhere. Whole legs can be lost this way.
  3. Place one foot off the curb and onto the road. I find it helpful at this point to pray to the Traffic God and Road Demigod to grant us safe passage. My favorite prayer is to utter under my breath “Please don’t kill me” as many times as it takes to get across the street.
  4. Place the other foot onto the road and prepare to push off using that foot . . .
  5. Go! And keep praying. If at this point you feel the public attention waning, feel free to flail your arms or hold them out in deference to the waiting motorists. This will reinforce the response evoked in step 1.

In homage to step 3 I have named this maneuver the ‘Please Don’t Kill Me Maneuver.’ There are a few notes to keep in mind.

The first is, never panic while crossing the road using this method. If you appear nervous or weak the drivers will sense it and pounce. Do as one would do in a mountain lion attack: Yell and make yourself appear as big as possible. This will scare off the smaller ones and unnerve the alphas.

The second is actually serious: if you fall, don’t put on a show. Get up as quickly as you can and keep going. This isn’t to save face (you won’t, it’ll look ridiculous no matter what) but to get out of the intersection quickly.

Then again, if you’ve enchanted everyone properly, they’ll all just coo at you while you get back on your wheels. Babies falling down seems to enthrall humans more than anything.

After doing this for a week, I got tired of the unsteady feeling when skating down the little dip in the curb onto the street. So I devised a second ingenious method, which I dubbed the “Pretend You’re a Car” method. This is a technique pioneered and used by road bikers worldwide, which has been adapted as follows:

  1. Skate on the street, in the bike lane if you can.
  2. You are a car. Cars don’t feel. They are machines operated by humans.
  3. While waiting at an intersection for the light to change, concentrate on what you are. A Car. Cars don’t look around. They stare blankly ahead into the butts of all the other cars in front of them. If there are any bicyclists ahead of you, do this to them. Remember, they are cars too, so they won’t feel any emotion nor notice because they can’t see behind them.
    1. Don’t actually do this, I just thought it would be fun to write. Be reasonable on the road.
  4. When the light changes, cross the intersection. Celebrate because you are a car and don’t have to worry about curbs.

Utilizing this technique (especially in California and New York) you may be subjected to the all-common road rage. Remember, you are a car. Only humans feel the inflated self-importance that moves them to honk and yell at other humans piloting cars. The cars are merely there, the transitory shells through which humans move and bicker. You are a transitory shell. A visible ghost. You move gracefully through the shit while the world falls apart around you.

But feel free to flip those cats the bird cuz you don’t give a fuck.

Chemical Hood

This idea courtesy of the Oberlin College Biochem Department. They made a shirt with this on it. It’s the most badass major’s shirt ever. If you don’t get it, it’s a picture of chemical fume hoods, where Chemists and others spend a good deal of their life.

Because I’m still starting out with all of the above, I also picked up a road bike for cheep on Craigslist. Craig, you are a savior of poor students. For about $100 I got the bike, and fixed it up with a new rear tire and inner tube, plus a U-lock. I’d say that’s a good deal. The shifter is shot so I’m treating it as a single speed, which is working real nice except for yesterday when Lake Ave. in Altadena conquered me. That’s a hell of a grade to bike for an hour in the highest gear. In homage to something, I’ve named the bike Shing. Like the noise people make in their head when they pick up a real sharp knife. “Shing.”

Shing poses awkwardly against a pole

Shing poses bashfully against a lamp

Second things are second: Science

We’re working on these things called scientific visualizations. There’s a lot of voodoo magic terminology that gets thrown around with this stuff, but to me it’s just about making things visually intuitive. It’s about turning

Crappy Train Schedule

Crappy Train Schedule

into

Source: http://saezablog.wordpress.com/2011/09/18/train-system-in-japan/

Hours on the left column, minutes on the right. Together they form the Hour: Minute departure time.
Source: http://saezablog.wordpress.com/2011/09/18/train-system-in-japan/

In other words, science experiments spit out a lot of data which is tough to interpret. We’re the A-Team behind making all the numbers look like they make sense. But without lying. And since humans are damn good at seeing things, that’s the way we go. It’s interesting to think about translating data into something else, like music, or into words, which is also very common. It’s usually called music and poetry.

What I dig about this DataViz stuff is that it opens up the creative, artsy side of compsci. Having spent most of my life being very rule-oriented and ‘proper’ (and having opened up lately thanks to Capoeira and friends), I really like the opportunity to approach a problem that requires more than algorithms to solve. Something that doesn’t have a ‘correct’ answer.

It’s also interesting to look at the world through the lens of DataViz – which is honestly just looking critically at the world. I only say DataViz because that’s the area that got me to look at the world critically. But the point is, it’s really interesting to look around and realize that none of the things in the world are necessarily ‘right’. That’s just how they were done. Can openers are the way they are because that’s how the lineage of can-opener designers, engineers, producers, found was both useable and cost-effective. And when someone comes along and creates a can opener that cuts along the side of the lid so that it can be replaced if you don’t use all the contents (Oxo’s ‘Smooth-Edge Can Opener‘), it makes me blink. Because, why didn’t we do that before? And now that you think about it, how does the original design make sense?

When it comes to DataViz, we take this idea and apply it to numbers and graphs and all that nonsense. Is that really the best way to show the data? Is there a way that’s more intuitive? This is a big deal in the science world because most researchers want to show their findings as quickly as possible with minimal overhead. That usually means plotting things in Mathematica or Matlab. But just because the tools are readily available, doesn’t mean they illustrate what’s going on well. So we run into a lot of interesting lifehacks that researchers use to make their graphs carry more meaning, within the limitations of the tools they know how to use. Maybe use color to denote that something is moving toward or away from you, or different plots of the same thing [Warning: Brains] so you can cross-reference what’s actually happening. This happens a lot with 3D objects; people typically look at one thing from some different sides to see what’s up.

So we’re working on three projects, which We’ve tagged: Fluids, Brains, and MBSE. Fluids referring to turbulent flow over a wall, a project on which I’m the lead programmer, Brains referring to Resting-State Networks in the brain, and MBSE referring to systems engineering. It’s a tough set of problems and we’re chugging away on them, already approaching the fourth week. This internship is pretty rad in that we’re writing the book as we go – as in, we’re creating our own project frameworks, structures, deadlines, etc. I hear it’s a lot like grad school, where you pick a problem, have an advisor or two, and go! Also we’re working at NASA which is unbelievable for so many reasons. One is because it’s been my dream since I was |—| <– that little, and two because it’s literally like being in a scene out of a scifi film. I don’t know how much I’m allowed to describe so let me illustrate analogously.

So this coming week we’ll be doing proofs-of-concept and physical prototypes of the programs. This is one of my favorite things to do because I get to code up a bunch of small throwaway programs to test out different languages and packages. It’s literally like being a kid in a toystore, except all the toys are free and come with infuriatingly long instruction manuals. Documentation aside, it’s very fun. So that’s what’s coming up for the week. But right now it’s the day after LIBERATION DAY!!!!!!!! so it’s no time to be thinking about that.

Today I’m planning to check out MolyJam with some of the Caltech Kids. Maybe it’ll be hot. Maybe it’ll be not. We’ll see as the day unfolds. I also made a big pot of soup today out of all kinds of farmer’s market veggies and a coconut. Soup is fun to make. Hopefully tomorrow will be a beach day.

Okay there’s lots more things to write about but they’ll have to come in later installments because I’m bored. Time to go play.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s