Student Life
Who knew a fellowship could keep a person so busy? When I re-started this blog, I thought I'd easily be able to post several times a week. And at first I could.
Then classes started.
Technically, as an auditor, I'm not required to do the homework (or "problem sets" as it's called here), take quizzes, or tests. I suppose I don't even have to do the readings or show up, but then what would be the point of "taking" the class?
But as a matter of course, I do do the problem sets. I read the material. I attend class. And as a result, I've been way busier than I expected. (To be clear, I'm not a masochist. I'm not taking the quizzes or tests.) And I'm doing this for the five or so classes I'm currently taking. Mostly.
Here's a quick rundown of each course and what I've taken away so far:
Genetics
An introduction to genetics class, this course meets for an hour three times a week in a medium-size lecture hall. It is the kind of science class I think most people think of when they think of college science lectures. The blackboards sit three deep and rest on rails allowing the two professors to raise and lower them at will.
It started with simple Mendelian genetics and quickly delved into the mathematics of inheritance traits, including those that are caused by multiple genes.
The first class was fun. I understood everything (and even felt kind of smart for knowing stuff already). Toward the end of class, the professor brought out vials of fruit flies for a demonstration. One vial contained normal fruit flies. The second contained flies with a genetic mutation that caused them to be temporarily paralyzed when exposed to heat. Using an overhead projector, he demonstrated the effect to the oohs and ahhs of the students. Maybe not fun for the flies, but definitely fun for us.
Truth is, biology was one of my weakest subjects in high school and this course is a great reminder why. For starters, as Steve Martin once joked about the French language, "It's like they have a different word for everything."
Seriously, as if the science isn't enough of a hurdle, you also have to learn an entire vocabulary specific to the field. At first, I was nodding along. But soon I was scratching my head trying to remember the different between meiosis and mitosis and the rest.
Yet, when I scan the room, I see the other students — few old enough to legally drink — nodding along. I'm glad they get it and I look forward to their discoveries and work in the years ahead.
To be fair, my goal with the class was to better understand the broad concepts around genetics and especially the work being done around genetic manipulation. The early lectures are more foundational and are setting the students up for more detailed biology classes to come. Since the topics I'm more interested in are covered later in the course, I've pulled back from this class a bit with the intent of returning to it later in the semester.
Earth Science, Energy and the Environment
Unlike genetics, this class is intimate. With fewer than a dozen students, the professor knows us each by name and engages us in more of a dialog than a pure lecture.
We started at the beginning. Literally. The Big Bang. The formation of the Earth. And that leads to questions.
Why do we think we know the Earth is 4.6 billion years old? (Aging old rocks.)
Why do we think we know the Earth has a liquid iron core? (Average density and composition of asteroids, for starters.)
What should we do about global warming? (That's a big question...)
The class is fascinating and delightful. One day the professor passed around asteroids, moon rocks, and a piece of Mars that fell to Earth after an asteroid slammed into the red planet.
Another day, we passed around a piece of shale that was about 20 percent oil, not that you could tell from the feel or smell.
Of course some days are less delightful than others. For example, the day that was devoted to working through the differential equations that determine the heat and pressure of rocks at certain depths. Oy.
Overall, though, the class is like Nemo looking at the code in the Matrix. Through an understanding of the planet's geology, we see how everything comes together and with that can solve all kinds of puzzles and problems and perhaps even work out a solution to climate change. Maybe.
Also cool: the professor is working with NASA on an Earth-monitoring satellite. You know, side gig.
Extrasolar Planets: Physics and Detection Techniques
In 2013, Sara Seager was awarded a MacArthur "Genius Grant." She had made a name for herself for her work on exoplanets, including analysis of their atmospheres. So it's pretty neat to be sitting in a small room with her and a dozen undergraduate students learning how to find and analyze planets orbiting stars millions of lightyears away from Earth.
There are two main ways exoplanets are detected. One method is called radial velocity and essentially boils down to this: planets cause the stars they orbit to wobble. By measuring the wobble, we can measure the planets. The second method is by looking at a star's light and measuring dips in the brightness of the star caused by a planet passing between the star and Earth.
Both methods seem straightforward, and neither is.
For radial velocity, it might be relatively simple if there was just one planet orbiting a star. The wobble would be entirely due to that single heavenly body. But of course it's never that simple.
Consider our own solar system. We have eight planets, plus dwarf planets, asteroids, comets and other objects all orbiting the sun. Each object exerts pull on the sun, causing it to wobble in all kinds of ways. Parsing that data to determine the number, size and location of the planets is anything but simple. Luckily, brilliant people build tools to help figure it all out. Still, someone had to work out the math first.
Furthermore, the sensitivity of instruments required to measure the wobble is extraordinary. How extraordinary? Scientists are trying to detect a wobble that can be measured to within four atoms on an Earth-based telescope. To do so requires incredible engineering to isolate any and all motion and to account for the atmosphere, people walking, changes in temperature and pretty much anything else you can think of.
As for transiting planets, imagine your favorite reading place. You're sitting in a comfy chair with a soft-white lamp over your shoulder as you page through a mystery or biography or romance novel. A gnat flies between you and the lamp. Did you notice the light dim?
Of course not, but it did. A few fewer photons reached your book as the gnat zoomed past. And that's what happens with a transiting planet. By watching thousands of stars, scientists like Seager can see, in a few cases, planets passing in front of their stars, dimming their light just a little. This process only works with a small percentage of stars — after all, the planets have to transit between the star and Earth (some might orbit on a completely different plane), and while we are observing it — but with so many stars, we have observed thousands of transits.
Amazingly, what can be discerned from transits goes well beyond what seems possible to the uninitiated. Scientists can determine the radius of the planet and the distance from its star just from the transit (and knowledge of the star's size and mass). As a result, Seager and others have discovered a wide range of planets in all sorts of orbits. For example, some planets are as big (relative to their stars) as Jupiter is to the sun, but with an orbit inside that of Mercury. These so-called "hot Jupiters" whizz around their stars with astonishing speed.
And there's so much more, but it'll have to wait. Oh, and Seager is involved in the NASA Tess project. It's almost like everyone has a NASA project around here.
CS50: Introduction to Computer Science
Among my great shames is that I've never gotten very good at programming. Oh, I know enough to be dangerous and to, generally, understand people's code, but I'm just a punter (as my Australian wife would say) when it comes to really knowing executable code. So I figured the fellowship would be a great opportunity to fix that glaring hole in my brain (one of many, to be sure).
Cs50 is a famous Harvard class — and a gargantuan undertaking, as I previously wrote about. With more than 100 staff, it's a small company unto itself.
At first I was into it. The so-called “easy” problem sets were fairly easy. The hard ones were more challenging, of course. I'd spend a few hours on them, get frustrated, take a walk, and then work out the answer. Yay me!
Then I started attending the staff-led sessions. The first one was headed by a brilliant young woman who had interned at Bank of America and Microsoft. She is 19.
As she delved into memory allocation (malloc, in C parlance), the memory buffer overruns she described on the board started to manifest themselves in my brain and I crashed.
I re-watched the videos. Studied the online lecture notes. Read through the session slides. Pored over discussion forums.
Still, the next problem set had me stumped. I spent an entire weekend trying to work through the challenge. One night I stayed up to 3 in the morning. I made progress, but still fell short.
Finally my senses got the better of me. Why was I killing myself trying to learn C? That's not my future.
So now I watch the videos and absorb away the broad concepts. So much better.
CS with Python
As I stood down from CS50, I ramped up my MIT Python course. Among the things I hadn't fully realized was the degree to which computer programming is now de rigueur for all scientists. Just recently Sara Seager spoke about the need to write in Python to work through exoplanet calculations.
Whereas CS50 starts with low-level computer science as an on-ramp for those who might end up making the next iOS, CS with Python is aimed more squarely at my level: using programming to solve problems in other fields.
As I previously intimated, my programming skills are slow-pitch softball in a hardball world. I get the ideas, but I'm slow and crude. So this class ought to be just right. Especially because it's taught by two professors who started programming when computer bugs were actual bugs. (See the featured image on this post — it's the first programmable computer and it's in the Harvard science building. And it did have an actual bug in it.)
Although I've fallen behind in the course — mostly due to recent travels and a temporary focus on Indians baseball (I don't want to talk about it) — I'm looking forward to catching up and upping my game.
Applied Data Visualization
A recent addition to my course load as been a Media Lab class in Applied Data Visualizations. A three-hour class on Friday afternoons, the course is perhaps the most engaging yet — thanks to the teaching assistants who lead it. With students from a wide swatch of colleges in the area — that is, not just MIT — the course is a fun hands-on workshop focused on using D3.
Despite all my work in data journalism and interactive storytelling, I've never spent much time in D3 and this course gives me the structured excuse to correct that. It also is introducing me to people with a passion for data visualizations and a different perspective on what can and should be done.
In addition, it has given me insights into how I can better teach my own classes (should I be lucky enough to teach again). And in that way, it is the perfect course — inspiring, insightful, and useful.
Coursera
Finally, and perhaps most foolishly, I've also been taking Coursera classes. It seems ridiculous — here I am within easy biking distance of Harvard and MIT and I'm taking online classes? But I figured with more time on my hands, I could and should fill in other gaps or re-up my knowledge in areas that have atrophied. Statistics, R, Data Science, Illustrator, Lightroom — why not?
Then there are the classes I considered taking but decided not to.
How to Make (Almost) Anything
As one student said on the elevator heading to this class, "This is the prototypical MIT class." And it really is. The class is basically a course in how to fabricate whatever you want. That means making devices from scratch, including circuit boards, and programming them. Run out of a "Fab Lab," which has since been replicated around the world, the class is the "maker" revolution at its core. Imagine a world where, if you can think it, you can make it. That's this class.
You can browse the course website (take a look at prior years and what they made), but I like these two examples: An alarm clock that pulls your blanket off, and a scream capture-and-release system.
VR/AR Immersion
In a later post I'll talk about the project I am working on during this fellowship, but suffice it to say it involves augmented reality. So when I saw this class, I thought it was the perfect one for me. And maybe it is. However, it conflicted with two others I was enjoying and so I was loathe to change my schedule.
In any event, this class is a collaboration between the Berklee School of Music and MIT. Musicians and engineers are working together to build virtual reality and augmented reality apps that feature sound and music. These apps can range from projects focused on how to compose sound and music for virtual environments to using VR or AR (i.e., how do you make sure your experience is aurally appropriate to the space you are in) to interacting with sound- and music-producing objects in a virtual world.
Among the striking things about the class was the plan to bring in venture capitalists toward the end to work with students on building companies around their creations. Truly a different world from my college experience.
Science and Technology in Domestic and International Policy
Technically I'm signed up to audit this class at the Harvard Kennedy School of Government taught by President Obama's science advisor, John Holdren. In reality, I attended one class and haven't been able to make it back since, due to scheduling conflicts.
Despite ignoring one of the basic lessons about how not to use PowerPoint (don't just read off the screen), having the opportunity to tap into Holdren's deep and extensive knowledge is a gift. His discursive tales of the intersection of science and public policy were illuminating and, at times, funny. In one story, he explained that Lawrence Livermore National Laboratory was created, in part, to keep the much-reviled Edward Teller busy and away from everyone else. Heh.
One of the truisms about the fellowship is that you simply cannot do everything. And in this case, I've been unable to make other lectures. Perhaps I'll get to more in the future. I hope so. In the mean time, I've invited Holdren to dinner with the fellows, several of whom are in his class (and attend regularly). I'm still waiting for his response.
So, that's some of what I've been doing instead of writing updates.