A plan for clean air

In this city, the pollution is pretty bad. There are just so many cars and motor scooters going around the city that the air quality ranges from decent to terrible. The problem is endemic to dense cities: with streets so close to everything else, there isn't enough room for the ghastly exhaust from highways and streets to dissipate before it hits your precious lungs. The good news is that we can lick this problem in about a decade.

That's an optimistic prediction, but I really think we can do it. If not in ten years, then in fifteen or twenty. But it will happen, and we'll live to see it. And smell it.

The first ingredient: some new technology

The first big thing we'll need is plug-in hybrid cars. These are like regular hybrid cars, with an engine and a battery that can power the car for a while, with one little difference: the battery is more powerful, and you can plug it into an electrical outlet overnight to charge it up. The result: effectively an electric car for city driving, which miraculously turns into a gasoline-powered car when you drive for longer than the battery can support. Powering a car on electricity costs (in California, where the electricity can be pricy) about a fourth of what it would cost to run on gasoline. That's some pretty impressive savings, especially considering the fact that oil prices just keep going up but our electrical power costs will be more stable for the next few decades at least.

Imagine one of those pollution-heavy streets minus the pollution. And imagine that the cars are dramatically quieter than they are now, so there's less noise. If plug-in hybrid cars become common, then that's exactly what will happen.

This bit of technological speculation is not as radical as it sounds. The major automobile companies have been announcing production dates for their plug-in hybrids. The Chevrolet Volt is scheduled to begin production in 2010, with an all-electric range of 40 miles. They've designed the car so that it can switch to other fuels or go all-electric as battery technology improves. GM is also going to be making a plug-in version of their Greenline Vue hybrid for release in 2009, with a smaller electric range of 10 miles. Ford is lagging behind, saying that they'll do it in about five years once battery technology gets better. Aptera Motors is going to start production in late 2008 of the most awesome-looking car in the world, which will get between 130 and 300 miles per gallon depending on how it's used. It will be able to go for 120 miles before the engine kicks in, so it's also going to be sold in an all-electric configuration. It looks odd, but they're selling it for $27,000 to $30,000 with a long list of pre-orders, and that gas mileage is bound to look pretty good to people suffering from high petrol prices.

It's not just cars, either; electric motorcycles will only become more practical as battery technology improves. I realize that a lot of motorcycle owners in America probably ride it for the noise and power and would be affronted by the notion of switching to an electric motorcycle. But in countries like Taiwan, you see motorcycles everywhere because they're so useful. There, where pollution from motorcycles is more important, people tend to use motorcycles more as practical transportation devices than as toys.

The second ingredient: some of that hippie "green power" we hear so much about

Batteries just store energy. That energy has to come from somewhere, and right now that usually means coal. It would be nice if we could improve the situation a bit. Guess what: we can!

Environmentalists are always talking about how "renewable" energy (wind, solar, tidal, etc.) can be a huge thing if we just paid the premium to implement it. Well, guess what: that premium isn't so big anymore.

Photovoltaic panels (which convert sunlight directly into electricity) get cheaper to produce every year. The old expensive things that people had in the 1970s are a thing of the past. Modern solar panels cost less, use much less silicon, and can often be printed in rolls like sheets of paper.

Google has found it profitable to cover their headquarters with solar panels recently, and their plan makes sense. Not only do they get warm happy feelings and good PR, but it's also insurance: they're saving money on electricity now, and in the future as demand for electricity continues to rise, they'll still be paying the same amount for their solar power. More businesses should follow suit -- and so should individuals. We've reached the point where installing solar panels atop a new house in sunny climates really does make economic sense. Especially in places like California, with power woes and real estate so expensive that the cost of solar panels is lost in the noise, people need to supplement their homes with solar.

Wind power is also showing some promise. It's competitive in cost with solar power, but requires bigger blocks of investment. The same goes for tidal power, and some of the other "renewable energy" technologies.

One big problem with this green power is that it can't provide a constant supply of power. Solar power only pays out when the sun is shining, windmills only turn when the wind is blowing, and so on. Good news: car batteries can soak up excess power when it's cheap. In fact, green power and plug-in hybrid cars have other good effects on our power grid that aren't immediately obvious, and aren't talked about much in the news.

Power use varies over the course of every day. The coal plants that make up the bulk of our power supply (and the nuke plants than can do it cleaner) are hard to start up and shut down. You can run the generators at higher or lower capacity, but there's an optimal range of power output where the generators are most efficient. Outside that range, things go progressively more and more wrong. So we use other ways of meeting the peak power demands around noon, like running hydroelectric power in reverse at night, or using natural gas peaker plants that can be turned on in a hurry.

One of the things that's pretty nice about solar power is that it tends to have its peak power output at the same time the peak demand hits: the middle of the day. This has the effect of smoothing out the power demands on the rest of the grid. Since we can charge up electric cars when power is plentiful (usually at night), this will also smoothe out the demand.

Having a smoother, more constant demand for electricity has lots of beneficial effects. We won't need to worry so much about intermittant problems with overloaded power lines, and more importantly we can meet more of our energy needs with the more efficient baseline sources. That buys us more improvement than just the difference between the efficiency of a car engine versus the electrical grid. I'm not sure how much improvement, but it should be significant.

The third ingredient: better batteries (mostly optional)

One big problem with electric cars and plug-in hybrids is that our batteries aren't cool enough. Sure, they mostly work, and all the miracles that I described above are possible with current battery technology, so I should be more appreciative -- but batteries seriously need to get better. They add thousands of dollars to the cost of hybrid cars, and more to plug-in hybrids and fully electric cars. They run out too fast. Some kinds wear out after several years, or too many cycles of charging and discharging. It can be hard to draw a lot of current from a battery all of a sudden for a sudden burst of acceleration, which is why electric cars have (until recently) been sluggish laggards. The usual way of dealing with this is to add supercapacitors, which also cost money, but will give snappier performance and recharge faster -- good for hybrids that want to recapture energy when they brake.

The battery situation is changing for the better. All of the technology I'm going to talk about in this section is still in the research phase: we have prototypes, but moving it into mass production will take time and (relatively small) investments from car companies. Still, you can be pretty confident that some of this will be commercially available in the near future. Remember: this is just a sample of the exciting new developments that have been happening recently. There's a lot more.

First, we have a way of increasing the battery capacity of the batteries we put in cars: Stanford university's nanowire lithium-ion batteries. These use essentially the same mechanism as existing lithium-ion batteries, but part of the structure has been changed to use a particularly hairy kind of silicon structure. This lets the batteries store 10 times as much energy in the same space. They still need to make sure the battery lasts long enough, that it's rugged enough for a car, and that it's engineer-friendly. They also need to get mass-production going, but once they do it should be about the same cost as existing lithium-ion batteries.

Some other scientists have taken advantage of the cool properties of carbon nanotubes to make some particularly unusual batteries. They're very light, they're thin, they're flexible, and they're biodegradable. They're paper. More precisely, they're paper infused with carbon nanotubes in a particular arrangement. They can behave as both a lithium-ion battery and as a supercapacitor. Make a battery with these, and you can get the long-term power draw you expect from conventional batteries, and the short bursts of energy for which supercapacitors are so useful. Right now, they're trying to figure out how to cheaply print this stuff into giant rolls of paper. Then they can stack the paper to make cheap batteries.

The third fascinating new thing in better batteries isn't really a battery at all: ridiculously advanced supercapacitors. Some people are calling them "ultracapacitors" in an attempt to make them sound almost as cool as they actually are. They store energy in an electric field, and they're increasing in energy density. A company called EEStor claims that it will soon be selling ultracapacitors with a higher energy density than conventional lithium-ion batteries. This doesn't sound overwhelmingly exciting -- after all, I just finished talking about new batteries with several times that energy density -- but capacitors can go through vastly many more charge-discharge cycles without dying, they're environmentally friendly, and they can charge and discharge mch faster. The company is even claiming "4-6 minute recharging", if you have beefy electrical cords and you can provide a ridiculous amount of current without blowing a fuse. (They don't talk about that last part much. Oh well. It's still some awfully nice technology, and gas stations could give you a quick recharge with some special equipment.)

So it looks like the hybrid cars (and even the all-electric cars) will soon have very nice performance -- and they'll have it cheaper than today. Which brings us to the last topic: money.

The final ingredient: economic incentives.

Look around any American city and count the number of unnecessary SUVs, and it will quickly become obvious that the desire to be environmentally friendly and socially conscientious doesn't pull much weight around here. Before more fuel-efficient cars will be widely adopted, we're going to need to make it an obviously good deal financially. (Or a hotly desired status symbol. But I'm a straightforward guy, with no gift for the mildly disgusting business of adversising, so I'll stick to talking about economic incentives.)

A lot of the really hard work here is being done for us by the economics of oil: gasoline prices are painfully high, and they'll get higher. People will happily switch to something with better gas mileage if it has good handling and they don't pay too much extra for it. The plug-in hybrids that will be coming out soon meet both requirements, and they'll only get better as the technology improves.

We can make the gas prices even higher by raising gas taxes. We could use the extra money to subsidize the purchase of more fuel-efficient cars. This scheme has some problems, though. For one thing, a lot of our society's infrastructure needs gasoline. Raising gas taxes will have ripple effects through our economy, making other goods more expensive. Do we really need to penalize diesel-electric locomotives for using fuel? Or buses? Or police cars? And gas taxes are regressive, affecting the poor more than the rich. If you need to get to work or take your kids to school, and you can't afford a better car and public transit isn't a viable option, then you're pretty much stuck paying for too much gas.

One answer to this is making important services and lower-income people exempt from some of the gas taxes. The potential for political lobbying and corruption is horrendous, of course, but with only average amounts of luck we should be able to get a system that kind of mostly works.

We also face another problem in trying to get people to be more fuel-efficient: a few lousy cars use a remarkably large amount of gas. And for some reason their owners don't get with the program and get better cars. If we offer incentives to replace those old fuel hogs with newer vehicles (and later maybe even ban them), they will eventually go away.

If my fellow environmentalists want to lobby about something, this would be a good thing to start talking about. I think we have the political ability to make this happen in some key states like California, and that's too big a market for the car makers to ignore.

Conclusion: This is going to happen!

Even if we somehow don't get the shiny new technology that we're going to have soon, even if we don't get the help of the politicians, we're going to convert to plug-in hybrids and electric cars on a wide scale in the coming years. Ten to fifteen years from now, I fully expect to be able to breathe much cleaner air even in the heart of a big city. It'll be cheaper in the long run, too, and it will improve our economy. This is the kind of thing that everybody can get behind, liberal and conservative and libertarian and communist -- everybody.

Except the oil industry, of course. But what they think about it ultimately won't matter.

The Chinese language is evil and wants us to suffer

It turns out that the Chinese language is every bit as difficult to learn as people claim. It's not even the tones that are bothering me the most (I can hear the tones now, and reproduce them) nor is it the weird phonemes that most westerners can't pronounce (which I've figured out how to pronounce). No, the thing that trips me up is that everything sounds totally different from every other language I've heard. Take Spanish, for example: words like "comida" (food) and "avión" (airplane) sound like they're just a couple more oddly-pronounced English words to add to my vocabulary. A lot of Spanish words even have the common decency to be cognates of semi-obscure English words. When I hear "comida" I think "comestibles", and I'm home free! Even languages which aren't related to English, like Japanese, are dramatically easier to remember words for. More examples: "shojo" (girl) and "kaze" (wind, as in "kamikaze") certainly don't sound similar to English, but they're enough alike that you can get past the differences fairly easily and remember the words. Plus, Japanese has a fairly simple, easy romanization.

Chinese is different. There's a lot more syllable separation in Chinese, and the words just sound weird. I'd be teasing you if I didn't give examples. A phrase that means "eat something" sounds vaguely like "chr dong shi", all pronounced with a level tone. But my keyboard can't really get across the true pronunciation. For one thing, "chr" is a sound that most people in America have never, ever made. Imagine a typical growly "rrr" sound from deep in your throat, then move it way forward to the front of your mouth and cross it with an "oo" sound, with just a bit of "lll" thrown in for good measure. And stick a "ch" in front of it, and try to get a smooth transition from the "ch" to the weird buzzing that I just described. It's harder than it sounds, and it sounds pretty hard.

And then there's the subtle difference between "tsu" and "tzu", and the way the language makes you sound like you're speaking in a voice that's not your own. And the characters! The writing system was invented hundreds of years ago by people who figured that the logical thing to do was to make pictures to denote words or concepts. So they made pictograms. But they weren't content with that, since it has some obvious problems: how can you tell apart the pictures for dogs and wolves, for example? They make characters for numbers (like 一, 二, 三, 上, 下,etc.), and characters for sounds based off other words that sound similar (yes, ancient Chinese puns), and combined them in crazy ways, and then gradually changed them over time so that they look nothing like the original pictures. Check out how the character for "bird" changed from its birdlike origins to something that looks nothing like a bird:



And just look how crazy some of these Chinese characters are:

鄭眾

Fancy! And ubiquitous, too: Chinese civilization has been around for a long long time, and they've been using some kind of Chinese characters for almost as long.

But the ubiquity of Chinese characters has one problem, one hugely fatal problem: the Chinese writing system sucks. It's a lot more complicated than it needs to be, and the thousands of characters that everybody has to laboriously memorize offer precious little hint to their pronunciation. Sure, you might be able to read partial meaning of a symbol that you've forgotten, by recognizing some of its components. But you might run across a word that by rights you ought to know, like "knee", and have no idea what the character means. In America, well-educated people do not forget how to read and write "tin can". In China, it happens routinely.

It doesn't have to be this way! It's pretty easy to come up with a sensible alphabet for writing Chinese words. In Taiwan, children first learn something called the "Chinese Phonetic Alphabet", which consists of 37 simple symbols. Each symbol represents a single vowel or consonant, and together they denote every phoneme in Chinese. By composing two or three characters from this alphabet with a little slanted line or checkmark denoting tone, we can form syllables. It's shorter and faster to write than the full Chinese character set, and vastly easier to learn. Mainland China has something similar called Pinyin, which uses the Roman alphabet in some rather counterintuitive ways to get a mostly one-to-one mapping onto the Chinese language.

A lot of the younger generation has grown so accustomed to typing on a computer or mobile phone that they've almost forgotten how to write Chinese characters. See, on a computer keyboard there just isn't room for the thousands of symbols that everybody has to memorize. So on a keyboard you just type the word phonetically and choose from a list of symbols that are pronounced that way. This is much easier and more sensible than writing the characters by hand, and it's backwards compatible, so it's been widely adopted by "kids these days".

Now, everybody who's not hugely tradition-bound can see that the Chinese writing system is a problem. Mainland China has switched over to a somewhat less awful system called "Simplified Chinese", which has simpler characters, but still has the same old problems that we inherited from China's ancient history. They've also standardized on the Pinyin romanization system. That's more progress than Taiwan has managed to make; over here, people are still using the traditional Chinese characters, and there are a bunch of nonstandard romanization systems and a phonetic alphabet that no other country uses (and which children are supposed to stop using once they've learned how to obfuscate their writing like everybody else).

Whoever said that Chinese is a difficult language is absolutely right and deserves a biscuit.

Note: after I wrote this, I found an article explaining in detail and with fantastic wit why Chinese is so damn hard. That's its title. It does a better job that I ever could.

Food: what's good and what isn't in Taipei

Food is an important part of every single day. So here's a reference of where to get it and what to order, for my own benefit mostly.

Beef noodle soup is good, and can be found at the night market. Ask for something like "nio ro mien". There are tones in that phrase, but I can't be bothered to stick in the proper diacritics.

The school cafeterias have decent food in decent quantities for NT$50 (US$1.50) a meal. Go for the big cafeteria, because the smaller one has a smaller selection of generally slimy unpleasant food. Actually, "slimy unpleasant food" is a pretty major theme at the school's cafeterias, but in the bigger one you can usually avoid it. Go for fish, simple vegetable dishes like green beans, and fairly plain-looking meat. If you can find meat that's got weird red stuff on it, that's usually a safe choice. A bowl of rice and a bowl of that weird thin soup they have in the soup cauldrons will finish off the meal nicely. This is the staple of my diet.

There's a "noodle" restaurant that has the best dumpling soup I've ever had in Taiwan. Most places serve pretty nasty dumplings, but for NT$40-50 you can get a big bowl of soup with dumplings, scallions, lettuce, and broth. Hooray for broth! The location changes, I think: at lunch there's a stall at the market, but for dinner you have to go into the market, and take a left turn down a side street about halfway through. The restaurant will be on your right.

There's a bakery nearby called "Florida Bakery" which sells moderately priced baked goods. Actually, the baked goods are generally cheaper and better than the stuff you see sold at all the other stores, so it's a good place to pick up some bread. They sell a small loaf of dinner-roll type bread with a light dusting of sugar on top for NT$22 -- about 67 cents. It's a lot of bread for only 67 cents, and remarkably tasty. Avoid anything with "curry" in the name; it may look tempting, but it will only end in disappointment.

There are restaurants all over the place. Unfortunately they tend not to have anything written in English, nor even in the Roman alphabet that I like so much. If you go to a place with written menus, just look for something with a reasonable price (written in Arabic numerals, like "4" and "2") and put a single dash in the empty box to the right of the menu item. Then hope you made the right call and didn't just order a big plate of pig intestines and congealed pig's blood.

Speaking of blood, Taiwan has some weird foods that are probably better enjoyed by not eating them. Congealed goose or pig's blood is one of those foods. The taste is not really that bad but it's still pretty unpleasant, which combines with the texture and the fact that you're eating congealed blood to form a rather sub-par culinary experience. Stinky tofu is another of Taiwan's traditional delicacies, and it's better off avoided. Stinky tofu gets its name from the stench with which it befouls the air for meters downwind. The taste is earthy and surprisingly mild, but it's still not all that good -- and the smell lingers on your fingers for way too long afterward. Stinky tofu is on my list of things to erase from existance if I ever get the ability to do so with impunity.

If you go to a traditional Taiwanese restaurant, they may try to serve you rice porridge with sweet potato. It's bland and hard to eat with chopsticks (because it's porridge), and doesn't do a good job as a rice bowl for your meal. Steer clear of it.

Fish ball soup isn't as great as you would expect. Putting the balls on a stick and selling them at 7/11 doesn't make them any better. Aged hard-boiled eggs have a texture that's somewhere between eggs and leather, and they induce a surprising amount of queasyness in the stomach.

What this blog is about

I am a junior in Electrical Engineering at Iowa State University. From January to August 2008, I will be studying at Tatung University in Taipei. This is obviously a big thing for me -- the sort of thing that ought to be recorded in a journal somewhere, especially a journal that friends and family can read -- so I started this blog.
Note that the idea, and much of the implementation, is a total rip-off of EE in Taipei, started by my friend Jon.