		Site Map \| Contact Us \| Search:

JOIN

The Planetary Society Weblog

By Emily Lakdawalla

Expect the Unexpected

Aug. 14, 2006 | 12:44 PDT | 19:44 UTC

Weblog Archive

by Andrew Westphal

Just a few days ago, on August 1, we launched Stardust@home -- a search, by thousands of volunteers, for the first contemporary interstellar dust particles ever returned to Earth for study. This "stardust" was captured in an aerogel collector that was carried by the Stardust spacecraft beyond the orbit of Mars and back to Earth, spending seven years in space and traveling over three billion miles. A re-entry capsule containing two aerogel collectors, one containing cometary dust and the other interstellar dust, parachuted down from space last January. At this very moment, hundreds of Stardust@home volunteers, our new colleagues, are searching digital micrographs of the Stardust collector for the tiny tracks of interstellar dust particles.

Click to enlarge >

A Particle Track in Aerogel
This is what a single image from a Stardust@home movie looks like on the "virtual microscope." Since no interstellar dust particles have ever been captured, these samples were created in a high energy particle accellerator. Credit: Regents of the University of California, Stardust@home

As we launched the project, our website said that everyone should expect both major and minor glitches. We also gave everyone three reminders:

-- If you're not having fun, you're not doing it right

-- Expect the unexpected

-- This is research: the outcome of this project is highly uncertain

In this series of blogs for The Planetary Society, I will focus on each of these three reminders. First, reminder #2: Expect the Unexpected.

As anyone who has been participating in Stardust@home knows, our expectation of major and minor glitches was borne out in spades. As we turned on the project at 11 a.m. Pacific time on August 1, we were immediately overwhelmed with traffic. We had done "load tests" of the Stardust@home website before, with the very kind help of members of The Planetary Society, so we thought that we were prepared to handle the traffic. How wrong we were! We spent the next few hours scrambling to add servers to handle the load, and we eventually did come back up, sluggishly, later in the day.

After this, things seemed to be running smoothly. But this was not to last. We were watching the Stardust@home forum and noticed that our colleagues were reporting strange images -- wedding photos, ballerinas, cars -- randomly inserted in our "focus movies". These images were strange simply because they were so normal -- they weren't risque and contained no hostile messages. Nevertheless, we assumed that we'd been hacked. We spent a couple of hours trying to figure out how this clever hacker had gotten into our system. Eventually, thanks to the detailed information posted on the forum by our colleagues, we realized that we hadn't been hacked after all, but that a subtle bug had been introduced in the image distribution software. The result was that rarely these random images were included in our focus movies. We hadn't caught it because the insertions were so uncommon.

We quickly learned a lesson from this. Not only are our volunteer colleagues all over the world critical to the successful search for interstellar dust particles, they are critical to discovering, diagnosing and fixing problems with the website and Virtual Microscope. This is truly an amazing and fantastic scientific collaboration. Our new colleagues, working all over the world, are collegial, productive, and helpful. It is genuinely fun working with them. They are doing essentially all of the work. We have been humbled!

Although in a very general way we expected glitches, we had no idea what would come up. Who could have predicted ballerinas? But this is at the very nature of the scientific enterprise -- you never know what's going to happen next.

In fact, the possibility of finding something unexpected is what drives every scientist I know. Here is an example. In 1936, Carl Anderson at Caltech discovered a new subatomic particle that resembled the electron but was about 200 times heavier. This particle is now called the "muon", and we now recognize it to be one of the handful of fundamental particles -- the electron is another -- in the so-called Standard Model of particle physics. (The Standard Model is our picture of the structure of matter at the very smallest scales that we can probe.) This discovery came completely out of left field. A prominent physicist of the day named I. I. Rabi, reportedly on the way to a Chinese restaurant, quipped "Who ordered that?" But this unexpected discovery was not anomalous, but absolutely typical of progress in the field. In a wonderful letter to Physics Today a few years ago (Physics Today, July 2000, p. 15), Harry Lipkin pointed out that almost all discoveries in particle physics (the strange and charm quarks, the tau lepton -- a cousin of the electron and the muon -- CP violation, and many more) were completely unexpected -- that is, they were of the "who-ordered-that" variety.

Unfortunately, the way that most people think about science bears little resemblance to the way it is actually done. We are often taught that the Scientific Process consists of forming hypotheses and conducting experiments to test them. To reinforce this picture, in science classes we generally follow fixed recipes (misleadingly called "experiments"). We get an "A" if we get the result expected by the teacher. What could be more boring? To be sure, sometimes science does consist of hypothesis testing, but much of science does not proceed this way at all. In some fields, like Astronomy, hypothesis-testing is practically unknown. To pick a trivial (but typical) example: Galileo was not testing any hypotheses at all when he looked at Jupiter for the first time with a telescope and discovered four new worlds -- Io, Callisto, Ganymede, and Europa. More recently, astronomers were not testing any hypotheses when they discovered the mysterious "dark energy." They were simply exploring.

Click to enlarge >

Jupiter system montage
Galileo Galilei first observed Jupiter's four planet-size moons -- Io (upper left), Europa (center), Ganymede (lower center), and Callisto (lower right) -- in 1610. These images of the Galilean satellites were photographed in early March 1979 by Voyager 1 and assembled into this collage. The moons and planet are not shown to scale. Credit: NASA / JPL

But there are two other problems with presenting science in this recipe-like fashion. First, the challenging -- and fun -- part of science is figuring out how to make a good experiment or build a good instrument. If the experiment or instrument is handed out on a silver platter, you've missed half the fun! Just following a recipe that someone else wrote is like watching only the closing credits of a two-hour action movie. If that were what science were about, honestly, I'd be happier flipping burgers.

What about the other half of the fun? This is the possibility of new discovery. If you only get an "A" when you get the expected result, there is little room for new discovery. No wonder many people don't enjoy science in school!

When a practicing scientist does a new experiment, the outcome is highly uncertain. This is why it's called "research." This will be the subject of my next blog.