A surprising source of collective intelligence
I’m a bit late in writing about this, but it’s too cool to pass up. Last week, Google launched a new service, Google Flu Trends, that really demonstrates the power of new sources of information in the digital age. Flu Trends attempts to warn users of regional outbreaks of the flu. This enables hospitals, medical practitioners, and individuals to prepare. It’s not exactly a new idea. The C.D.C. publishes reports on outbreaks of influenza, based on data compiled from heath care providers. Another web service, whoissick.org, combines user-reported illnesses with Google maps to show you the various bugs circulating in your area. But Google Flu Trends may identify outbreaks more quickly because of the unique data source that it uses.
A Google team noticed that certain search terms, like “flu symptoms”, are much more common during flu season. Only logical, right? Google employees created a list of these types of searches, and compared the date/location of past searches with C.D.C. data on influenza trends. It turns out that the number of people with the flu and the number of these types of searches are closely related. This means that analyzing the numbers of influenza-related searches on Google should provide an estimate of the number of flu cases. By looking at IP addresses, specific regions can be isolated. This is a very nifty sort of collective intelligence based on data that is simply a by-product of Google’s primary function.
The sheer number of Google searches makes them an excellent source of collective intelligence. Nielson Online estimated that 4.8 billion searches were made using Google in September 2008 in the US alone. That’s roughly 160 million U.S. searches per day. In addition to the volume of data, search engine results are fascinating sources of information because of their timeliness. As the New York Times wrote:
[...] the data collected by search engines is particularly powerful, because the keywords and phrases that people type into them represent their most immediate intentions. People may search for “Kauai hotel” when they are planning a vacation and for “foreclosure” when they have trouble with their mortgage. Those queries express the world’s collective desires and needs, its wants and likes.
To me, this is a cool usage of data that most people don’t even realize they are generating. However, the implications for privacy are a little bit frightening. If you don’t believe me (and you have a Google account), check your web history. Google saves every search you make, along with any web pages you visit from the search results. A year’s worth of search data can create a surprisingly complete picture of a person’s life. Imagine having access to that data for every single user. I don’t want to be pessimistic about this. For the most part, I think the possibility of exciting and useful projects like Flu Trends greatly outweighs the potential hazards of this data. What do you think?
You can read the New York Times report on Google Flu Trends here. The official Google Blogs also covered it here. Statistics on the number of U.S. Google searches were pulled from the Nielson Online news release available here.
Thoughts on the Korean educational system
After a few weeks of teaching here, I’ve noticed some interesting things about the school and the educational system. My experience is limited to one middle school in a rural province, but these observations seem to hold true in other places, according to other teachers.
Korean students work hard.
Our kids are in school by 8:30am. Most of them have classes until 4:30pm. After school, they go home for dinner. Most of them spend the evenings at private academies, learning English, math, science, or music. If you’re on the streets between 9pm and 11pm, there are kids everywhere returning home, still in their school uniforms. They live and breathe school.
This is less than ideal, in my opinion. Erin’s co-teacher has kids, and says that they have no time to learn anything but school. They can’t cook, they don’t work part-time jobs, and they hardly see their family during the week. Hell, a lot of the kids are at academies during the weekend, too. I might be seeing things from an overly American perspective, but in my opinion, it’s hard to call that a childhood.
On the positive side, they take school seriously. Middle school boys are a handful no matter where they’re from, but I can’t help thinking that these kids get a lot more out of school than I ever did.
Grades/test scores matter, and they matter young.
In Korea, high schools are competitive. You don’t simply go to the school in your district. To be admitted to a well-ranked high school, you need good grades and test scores. I think they typically take an entrance examination as well. This means that young kids are seriously worried about their grades. The kids at my school are 14, 15, and 16 years old. (In Western ages, they would be 13, 14, and 15.) Even the 13-year-old kids take midterms that can determine their future. Poor grades in middle school, and you end up at a vocational school. It keeps them focused, but I imagine it’s rough on the late-bloomers.
Rote memorization is still the norm.
This is something that everyone says about Korean education. It’s important to note that I haven’t seen many classes, and this is a pretty broad generalization. However, I think it’s often true, especially with older teachers. Our co-teachers often stress group call-and-repeat exercises, rather than individual efforts. The kids have memorized responses to typical phrases, but often don’t understand them. It’s interesting to see.
As I said, this is often considered a common attribute of Korean education. In my opinion, this is overstated. I certainly can’t say that US middle school and high school stresses critical thinking. Even at university, it was often clear that students learned to fill in the blanks, rather than think. But overstated or not, education here is often a rote processes. I think this is due in part to the emphasis on good test scores. These kids test amazingly well, but it doesn’t necessarily indicate real learning.
Corporal punishment is still common.
Yes, some of the teachers hit the kids. I’d even say that a lot of teachers hit the kids. This one is changing quickly. I’ve heard that it’s illegal in Korea now. I’ve even heard that in the bigger cities, it’s not tolerated. Parents will complain or sue. In Gangneung, especially with older teachers, it happens. Usually it’s a token sort of punishment, a quick rap to get their attention. Sometimes it’s a bit more. I’ve seen a few teachers really smack the kids hard.
I have mixed feelings. I’m sure the educational purists are dead set against it. I’m sure there are other disciplinary measures that work as well or better. But here, it seems to be part of the culture. The kids obviously don’t like it, but they don’t seem traumatized. It seems to be commonly understood that it might happen if you really screw up. After it happens, they’re generally very well-behaved, and even good-natured about it. They don’t seem to hold it against the teacher at all.
As I said, it’s changing rapidly. The younger teachers don’t do it, and the kids know that it isn’t really accepted. If I had to venture a prediction, I’d say that this will completely disappear in the next 10 years.
Things really aren’t that different.
All-in-all, I’ve been more impressed by the similarities than the differences. Boys are boys, no matter where you go, and teachers seem to have a similar outlook. There’s the same sense of community (maybe commiseration) in the teacher’s lounge. Just like teachers back home, there’s always food being shared.
The more places I go, the more they all look the same.
A word about drug testing
I was just told by my co-teacher that today was exam training, so I don’t have to teach. A decent time to relate a recent experience that Erin and I had.
Our contract here required us to take a drug test, specifically a urine-sample based test. We were driven to the hospital last Wednesday, pissed in cups, and had blood drawn for HIV/AIDS. We weren’t concerned about passing either test.
Last Friday, a nurse came to school to get another urine sample from us. There was some problem with the previous test, but we were told not to worry. Maybe it was too diluted, I don’t know. We gave them another sample and didn’t think twice about it.
On Monday, our co-teacher pulled us aside and told us that we had failed the drug test, specifically the heroin/methadone test. She was shocked, and believed us when we said it must be a false positive.
Erin and I have both been taking a few different over-the-counter and prescription drugs. I’ve been using pepto-bismal for stomach problems, ibuprofen and acetaminophen for headaches, and diphenhydramine (benadryl, Sominex) for trouble sleeping. Erin has also been taking Sominex occasionally, and both of us had taken it Thursday night (before giving the second sample) We knew of another EPIK teacher that also failed his drug test after taking Tylenol PM (also diphenhydramine).
I decided it was time for some research. After an hour of searching medical journals, I had a few articles that mentioned false positives on the EMIT immunoassay test from diphenhydramine. I printed them out, showed them to our co-teacher, and arranged to take another test.
Our test is tomorrow (Friday) morning. I last took Sominex on Sunday night, so I should be in the clear. But needless to say, this entire escapade has been embarrassing. So here is some advice for anyone taking a drug test for employment:
- Do your homework! Find out what type of test they will be using, and study potential cross-reacting compounds (substances that could cause false-positives)
- Find out the next step after a positive test. For federal jobs in the US, the law mandates that positive EMIT tests be followed by a gas chromatography test, to rule out false-positives. This is obviously not the case in Korea, and could have caused serious problems for us. Insist on follow-up testing, if possible.
- Prepare your personal information. Make a list of everything you’ve taken in the past month. This should include prescription drugs, over-the-counter medications and any vitamins or supplements. Google-search for potential cross-reacting compounds. Bring the medications and prescriptions to the test with you. Print and bring any information relating to false-positives.
- Consider speaking to an attorney. I don’t have one and can’t afford one, so this was out of the question. Thankfully, our employer is willing to work with us. This won’t be the case everywhere, and a false-positive can cause serious professional and financial damage. It never hurts to know the law.
I didn’t have a strong opinion about workplace drug testing before this, but that has changed. I grant that it is necessary in certain situations, but policy needs to be crafted carefully. The real problem is that accurate tests like gas chromatography are expensive. Businesses are much more likely to spend a few dollars on something like EMIT than a couple hundred on a GC test. More likely than not, it will be on YOU to protect yourself.
A final note, watch for faulty information. A Google-search on these topics turns up thousands of shady pages written by stoners looking to pass a test. Most of the info is either hopeful or paranoid. Look for scholarly sources.
I’ll let you know how the next test turns out.
Robotic Monkey Arms
Man I love the title of this post. A couple days ago, the journal Nature published a report about monkeys that have learned to control a robotic arm with their thoughts, via a sensor implanted in their brains. Let me write that again. The monkeys can control robots with their brains. This is seriously cool stuff.
This is not entirely groundbreaking. Scientists have been working with this stuff for a while, and have been able to use brain waves to control simple things, like a cursor on a screen. I could be mistaken, but I think some of the technology started with the Air Force, which was researching thought-controlled computers for aircraft.
It sounds like this experiment was a serious step forward, however. You can read the NY Times article here, but I’ll summarize things quickly. The monkeys first learned to control the mechanical arm with a joystick, to get a feel for its movements. Then a chip was implanted on their their motor cortices , a portion of the brain that controls hand and arm movements. At first, the scientists used the computer to help the monkeys move the arm, but after a couple days, the animals could do it themselves.
Here’s a mental image: The monkey sits with its arms restrained. Using its thoughts, it causes the mechanical arm to reach out and grab food, which it moves to its mouth and eats. The researchers said that the animals even learned to adjust movements to compensate for sticky food.
This has obvious implications for prosthetics, but the long-term implications are so much bigger. Nearly everything humans do depends on our body’s ability to manipulate the world around us. Brain-technology interfaces could change the way we use computers, the way we drive, the way we work. More abstractly, they will change what it means to be human.
My friends will tell you that this is one of my favorite topics of discussion. After a few beers, I just won’t shut up about this stuff. The story of humanity is really the story of technology. From the club to the iPhone, we better ourselves by bettering our technology. A kid with a computer has access to information that people would have killed for only decades earlier.
Things are changing, however. Faster computers, nano- and bio-technology, brain-computer interfaces, better understanding of our our own biology…These things make new technology fundamentally different, in my opinion. We’re approaching a time when our technological potential outweighs our biological potential. When technology really makes us superhuman. Maybe this is the so-called Singularity…I don’t know. What I do know is that the future will bring closer integration between technology and our bodies and minds.
Sign me up for robotic monkey arms of my very own.
The Anthropocene Epoch
I heard an interesting interview on NPR while driving home today. Some scientists are arguing that we are entering a new geological epoch, characterized by human impact on the planet. The proposed name is the Anthropocene Epoch. The last 12,000 years have been classified as the Holocene Epoch.
The scientists interviewed gave some examples of how human actions were recorded geologically. With the widespread cultivation of arable lands, fossil records will show the shift from grasslands and forests to farm fields. Rising ocean temperature and acidity may threaten coral activity, which would also be noticeable in fossil records. Development has drastically changed the ways in which sediments are moved and deposited.
I’m not really writing this with any strong opinion, though some of the changes are clearly problematic. It’s just strange to think that the actions of our species will be recorded in the very geology of our planet. We’ve come a long way (for mostly-hairless monkeys).
Check the interview here.
