Stargate Universe logo

Stargate Universe didn’t get off to the greatest of starts, I know. The interpersonal drama was heavy-handed and poorly executed, the similarities to BSG were painfully obvious, and the self-contained plots weren’t sustaining interest all that well. But since the second half of season 1 started the Friday before last, I think things have gotten considerably better (whether this was the plan all along, or because Sci-Fi/Syfy/whatever paid attention to the complaints) and the series is heading in a positive direction.

No, it’s not the same Stargate we know and love. There’s no doubt it’s a very different kind of series: Stargate has always been about sci-fi action, but whereas SG-1 and Atlantis derived much of their appeal from their self-referential, lampshade-hanging humor, SGU has decided to eschew the comedy, and give everything a much darker tone. As a result, a lot of people complained that this somehow violated the “spirit” of Stargate. There may be something to that argument, but I also think you can’t just keep rehashing the same formula over and over again (that’s why the later series of Star Trek were so lousy), even if that’s what the fans want: eventually you’ll inevitably run out of ideas. As it is, that’s probably why Atlantis only lasted 5 seasons to SG-1‘s 10, they knew there wasn’t really anywhere else to go. I for one am pleased that they weren’t satisfied with another stale repetition.

And the other aspects of the show are improving too. There’s a intriguing plot arc now, involving an actual alien menace (in true Stargate tradition, the aliens even speak English, although not in the way and not for the reasons you might think), rather than the one-shot deals we had before. The tedious, angsty love triangles are still there, but they’ve been shoved into the background (not occupying more than 5 minutes of the last two episodes), at least for the time being. Instead, we now have people fighting for control of the ship, which makes for much more interesting watching. (Personally, I hope they get rid of the interpersonal drama altogether. Syfy already has one very good drama show right now, Caprica, which is just making SGU look all the worse by comparison. We’ll see how that pans out.). Characters are being fleshed out more: Dr. Rush is now a very polarizing enigma, with the show leaving it to you to decide whether he’s a backstabbing jerk or a brilliant mastermind who’s doing what’s ultimately best for everyone; Col. Young is no longer the de facto best choice for running the ship, considering the mistakes he’s made; and a lot of the assumptions we made about other characters are starting to look shaky too.

And to top it all off, the season 1.5 trailer for SGU appeared to have Michael Shanks in it. Daniel Jackson!

All I’m saying is, things may have gotten off slowly, but the pace is picking up, and I think the show is getting much more hate than it deserves. If you are among the people griping about it, try picking it back up (the last three episodes are all still on the show’s Hulu page), with an open mind that even though this isn’t the same Stargate, it could be a good show in its own right. You might be pleasantly surprised.


The New York Times has ran two pieces in the past two days relating to China and its currency. The first is that Timothy Geithner met with the Chinese vice prime minister, Wang Qishan, yesterday. Like so many diplomatic meetings, especially those involving China, both parties probably tried their hardest not to say anything of any real importance, so as not to cause any upsets, but it also points out that Geithner probably continued to nudge China to allow their currency to increase in value against the dollar. There are signs that may actually happen too: yesterday’s article mentioned a series of signs that China might allow the renminbi (or yuan, if you’re into that sorta’ thing) to appreciate.

Now, what exactly does this mean, and why is China doing it?

The thing to understand about exchange rates is that, most of the time, they are not set by some government authority or shadowy cartel of bankers, but simply by supply-and-demand, just like any other commodity. If the United States economy is doing well, investors all over the world want to put their money into American investments, and since they need dollars to do that, they have to exchange their local currency for dollars. Hence, demand for dollars increases relative to supply, and so the price of a dollar relative to other currencies (its exchange rate) strengthens. Likewise, if investors feel there are better opportunities elsewhere, they use their dollars to purchase other currencies, and the value decreases. Governments can take some action with regards to the value of their currency (last year, for example, Russia spent a great deal of its oil-money to prop up the value of the ruble, an effort that largely succeeded but cost Moscow a huge portion of its budget surplus), but not as much as much as they would probably like, and at any rate, the WTO looks askance at “currency manipulation” – unfair government interference on the value of a nation’s currency to gain an unfair trade advantage.

Many smaller economies, however, prefer to “peg” their currency to a larger one, usually the dollar or the Euro. There are a couple reasons for this: most notably that they prefer their currency to be shielded from moderate fluctuations in the market, which would have little effect on larger economies but could cause the value of their own currency to fluctuate wildly, which is obviously undesirable. China did this up until 2005, pegging their exchange rate at 8.27 yuan per dollar. Because of both pressure from foreign governments and the desire to make China a real player in the global currency market, China lifted the peg in 2005 and now values the yuan against a basket of currencies, but they have still endured a great deal of criticism that the yuan is being kept artificially weak.

Even if there’s no currency peg, there are a couple ways they could do this. One Chinese policy that is often cited as an artificial currency manipulation tool is their “purchase quota”: both Chinese and non-Chinese citizens can only change a total of $50,000 worth of yuan per year, in either direction. No one can sell more than $10,000 worth of yuan in a single day, or buy more than $500 worth (some exceptions are made for large corporations, obviously, since they have to deal with much greater currency amounts on a daily basis, but even these are viewed as needlessly strict). The criticism is that because these quotas impose hard limits on the number of yuan that can be bought and sold, they don’t allow its value to fluctuate with supply-and-demand.

This bugs other countries for a couple reasons: with a weak yuan, one dollar is able to buy many yuan, making Chinese imports cost less. The weaker the yuan, the less Chinese companies need to charge for their products, since it takes fewer dollars to net them the same amount of revenue. Likewise, foreign companies find it harder to do business in China, since Chinese customers have to spend more yuan to get the equivalent amount of dollars necessary to buy the product. There are complaints that China is performing this currency manipulation in order to give its industry and unfair advantage over those of other countries, both in China and elsewhere. Not only that, but it makes it more expensive for people to invest in China, which some people are saying is intended to unfairly limit control of Chinese companies to Chinese nationals.

However, currency manipulation is not the only explanation for these actions, there are some legitimate ones too, such as trying to prevent the yuan from getting too heavily involved in the “carry trade”. A carry trade is when people borrow money in one country, for the exclusive purpose of exchanging it for other currencies to borrow money in another country. Currencies with low interest rates and/or little economic growth are popular for carry trades. The thing about carry trades is that they are a double-edged sword: they can be very helpful for a country in some situations, and very damaging in others.

I’m going to use Japan as an example, because the yen is by far the most heavily carry-traded currency today. Ever since Japan’s recession in the 1990’s, the BOJ has set the interest rate of the yen extremely low (often zero), meaning for years it has been very cheap to borrow Japanese currency. Investors have taken advantage of this by borrowing tons of yen, converting it to other currencies, and then using that money to make investments in other countries, at less cost then it would take than to borrow the money in that currency directly. This has the net effect of weakening the yen against other countries (since everyone is trying to buy other currencies with their yen), which is a good thing for a highly export-driven country like Japan: the weaker the yen, the more yen foreign customers can buy with their money, and so the less that Japanese companies have to charge to sell their products, making them more competitive. Not only that, but a weaker yen makes the carry trade even more profitable for investors, so it encourages them to borrow and sell even more yen, a positive-feedback cycle.

When times are good, this is all and well, and the carry trade has been a great help to Japan for the last 20 years. The downside is when it backfires, it backfires hard: when there are no profitable investments to be found anywhere in the world, as was the case after the economic meltdown after the last 18 months, not only do investors stop borrowing yen, but they need to convert their money back into yen to pay back their Japanese loans. Hence, when there were few money-making opportunities to be had last year, everyone converted their overseas currency back into yen, driving the exchange rate up immensely (since summer 2007, where 1 dollar traded for ¥124, the rate plummeted to a low of ¥85 last November, and has since crept up to only ¥94). This is a major problem for Japan: with a stronger yen, Japanese companies have to either accept lower revenues (since the dollars of their customers translate to fewer yen), or increase their prices. This is why Japan has been hit doubly-hard by the recent economic turmoil.

Obviously China is not interested in seeing the same thing happen to them, which is in itself a perfectly legitimate reason for proper fiscal policy: the WTO doesn’t say that countries can’t be involved at all with their nation’s exchange rate, only that they cannot get involved so heavily that it “excessively distorts” what the value of that currency would be if only supply-and-demand economics were involved, with the aim of unfairly competing with other countries. The problem is that “excessively distorts” is a purely subjective measure. Countries like the United States believe China is excessively distorting the value of the yuan; China, of course, does not. Third parties are more mixed on the matter: The Economist says the renminbi is overvalued by about 20%, which is a significant amount, but nowhere near the 40% that the US tends to claim. China has slowly but steadily allowed the yuan to increase in value for the past few years, but the total change since 2005 has only amounted to about 16%.

If the Times is to be believed though, there’s about to be a significant policy shift, which may see the yuan increasing in value at a faster clip in the near future. The important thing to keep in mind here is that China is probably not doing this because of the pressure placed on them by the West. As the concept of “saving face” is extremely important to figures of authority in China, it is vital that they do not appear to their populace to be caving in to Western demands (the Chinese public is quick to notice and anger about such things). Instead, their primary motivation is probably economic: although the weak yuan has been a great help to China for the last couple decades, it’s starting to show some downsides too. Inflation in China has been higher than what the government would like it to be, and real-estate prices are skyrocketing. The problem is that with a weak yuan whose value is artificially constrained by policy, it is difficult for China’s central bank to manage these problems effectively; they are able to take much more action if the value of the currency is flexible with respect to the performance of the markets. That’s probably the real issue here.

At any rate, assuming the Times’ report is even accurate, the change is unlikely to be dramatic, at least for the time being.

Some people hate the iPhone. And that’s okay; we live in a free society after all, you’re welcome to hate whatever you want. Apple just made justifying your position a lot more difficult with their iPhone OS 4.0 presentation though. When I look at the list of reasons people give for disliking the iPhone, Apple sent a scythe through it today. To wit:

  • Multitasking – This was the big one, the primary functionality complaint. Steve admitted that Apple was a little behind in implementing this (no kidding), but said that the iPhone would do it the best. It looks like iPhone 4.0 will allow only specific services to run in the background, but since those services include the music player, GPS radio, data access, and a bunch of other things, that shouldn’t be an issue. They claim to have a “perfect freeze” implementation too – if an app doesn’t require background services (like most games), the iPhone can do a total freeze of the app’s state (so it uses 0 CPU cycles, and hence, no power), restoring it when the app is opened. I’ve always had my doubts about whether the impact of multitasking on a phone’s battery life is as severe as Apple claims- my Droid seems to handle it just fine- but this is a good thing.
  • Improved e-mail – Previous incarnations of the iPhone just didn’t have the same level of fine-grained control of your e-mail that Blackberries did, which is why the latter has remained the go-to device for serious e-mail addicts. 4.0 improves the situation dramatically though: there is now a “global inbox” to see the mail from all your accounts in one place, faster account switching so you don’t have to tap Back-Back-Back-Back-Back to get from one account to another, threaded mail viewing (finally!), and the ability for third-party apps to open attachments. All in all, this brings mail on the iPhone certainly ahead of that offered by my Droid, and probably on par with a Blackberry.
  • Enterprise features – The iPhone has ever-so-slowly been making inroads into corporate environments, but previous software versions were still lacking some vital features in terms of both usability and security. With 4.0 though, there is dramatically improved encryption capabilities (including allowing 3rd-party software access to the encryption APIs), distributing apps wirelessly with push notification for company updates (this one got huge applause from the audience), and SSL VPN support (VPN stands for virtual private network, and is a means of setting up an encrypted connection between you and a network which could be anywhere else in the world), which is a must in many corporate environments. IT departments are going to be hard-pressed to say no.

There were a few other announcements which were impressive but not earth-shattering. Like everyone else, they’ve tried to get into the “social networking” genre by including a unified gaming platform- perhaps the begging of an iPhone-Xbox-Live?- but what I really would’ve liked to see is the physical-buttons-add-on revealed in a patent a little while ago.

One announcement that’s not much from a usability perspective but huge from a business perspective is iAd, Apple’s unified advertising delivery API. Steve complained about the hodge-podge methods of delivering advertising content on the iPhone today, and thus unveiled a service where you could pay for ads through Apple, who would take care of delivery to apps that request advertising. The interesting thing is that Apple- not some other ad agency- is going to handle selling and targeting the ads… they are seriously moving in on Google’s turf now. Relations between Apple and Google, which have been bad these last few months, just got considerably worse.

There’s still no Flash, and obviously Apple’s stringent app-approval process ain’t going anywhere. But given that companies are rapidly redeploying their websites to work with HTML5, if for no other reason than to work on the iPad, it’s getting harder and harder to complain about the lack of Flash support. And with 150,000 apps to do pretty much anything you want, the second complaint doesn’t seem to be hitting as hard as it used to be either.

As usual, iPhone OS 4.0 is going to spend some time alone with developers, and will probably hit your grubby little paws in a few months… but only if you have an iPhone 3GS. Earlier iPhones and iPod touches either cannot get 4.0 at all, or get a degraded version without multitasking (which is the primary draw). This is probably going to irk a lot of people, although it makes sense from a hardware perspective. Both the original iPhone and the 3G are several years old, after all. The iPad’s getting it too, but not until the fall. This doesn’t make a whole of sense, but neither does buying an iPad without multitasking.

The DigiPen Institute of Technology is kind of like to game designers what Stanford is to computer scientists: other colleges make them, but these guys make the best. It was actually a team of students from DigiPen that made Narbacular Drop, the game that caught the attention of Valve, got the kids hired, and served as the spiritual successor to Portal. Yeah, that Portal. The students here know what they’re doing.

Well, this year DigiPen’s Student Showcase has produced another winner: Igneous. It’s a short-but-sweet high-speed platformer where you control a cubic idol-type thing and have to make your way throw a very active volcano, avoiding lava, pits, and the like. The only controls are to maneuver your idol directionally and make him jump, but he’s capable of jumping even on horizontal surfaces, so you can pull off some impressive wall jumps if you know what you’re doing.

It might be because my college schedule has left me without a lot of time to play games of late, but this was the most fun I’ve had from a game in a little while. It’s fast-paced, and demands your complete attention as you meander around collapsing pillars and giant lasers (don’t ask) at Mach 1. The visuals are quite stunning too, especially when you consider that the entire game was coded from scratch: no pre-built engine here. (The intense tiki music really adds to the atmosphere, too.)

When I was playing Igneous, I couldn’t help but think that this is what a Sonic the Hedgehog game is supposed to be: going at warp speed, with obstacles appearing nary a second before you have to dodge the, over and over, with one mistake sending you plummeting to your death. Obviously Sonic has not made the transition to 3D very gracefully; the faithful among us have been waiting for over a decade now for a 3D Sonic game that didn’t… well, suck, and so far have been disappointed. It’s a long shot, but when the kids who made Igneous get hired by a big game company (and they will get hired, I have no doubt about that), I have this secret hope that it’s Sonic Team that does it. Poor bastards could use the fresh blood.

The only downside to Igneous is that you need an Xbox 360 USB controller to really get the most out of it. You can play it on your keyboard, but it’s just not the same, and my cheapo Logitech USB controller wasn’t accepted. If you’ve got one lying around though, you owe it to yourself to spend the half-hour or so it will take to beat this game. Windows only.

Fenced in


Fences on the Cornell Collegetown bridge

You may have heard: Cornell has had a total of 6 suicides this year, 3 in the space of a single decidedly sobering month. Although we sort of have a reputation as a school where this is frequent, before this year it hadn’t happened at all since 2006, and only sporadically before that.

Clearly it was important for the University to take strong action to prevent this from happening again, and they definitely have. There’s been a great deal of outreach from the administration urging people to get help if they felt depressed or otherwise abnormal, and to do the same for their friends. This went as far as RAs going door-to-door in the dormitories asking people if they were OK. Pretty much everyone is in agreement that this was the right thing to do.

There’s less consensus about the University’s next step: the chain-link fences around the bridges above our iconic gorges. They’re ugly and they detract from the appearance of the area, no doubt about it. There is an argument to be made that they’re just a reminder of the horrible things that have happened, and aren’t improving anyone’s mood. I’ve heard a lot of people put forward the claim that, “If someone wants to kill himself, a fence isn’t going to do any good.”

This idea is innately intuitive, but like many innately intuitive ideas, it’s wrong. There have been numerous studies that have shown that although there are often warning signs for suicide, the act itself is usually impulsive, and people can be persuaded to call it off even with minor impedances, like a fence. While I agree that the fences shouldn’t be permanent (the higher-ups have already said that the fences are temporary while they find a better long-term solution), I think this is for the best. If it prevents even one more tragedy this year, it’s worth the unsightly appearance.

Lots of people formed very strong opinions about the iPad when it was first announced. As is typical for Apple products, these generally fell into one of two categories: people gushing that it was an incredible piece of magical technology that would change everything, and the usual contingent of Apple-basers with their laundry list of complaints- no open development, no multitasking (although that looks like it’s going to be fixed) etc.

Me, I decided to reserve judgement, because I could see a lot of legitimate points on both sides of the debate. On the one hand, I find the iPad much more appealing than any netbook. I hate netbooks and always have: their tiny screens and keyboards and underpowered hardware just get in the way when you’re trying to run a desktop operating system, be it Windows or Linux. The iPad has a small screen and keyboard too, but- and this is the key difference- its operating system was designed for that hardware from the start, and it seems to work great, from all the videos I’ve seen. (Side note: I saw some attempts, most notably Google’s Chrome OS to make netbook-friendly operating systems, and I was looking forward to these, but I think Apple beat them to the punch. On top of that, I just like tablets more than netbooks: it seems like you can do more with the technology.

On the other hand, I had some reservations. Not really about Apple’s walled-garden approach to app development, mind you: no matter how much I try to fulfill my “proper” role as a geek and hate what Apple’s doing, I just can’t get upset about the App Store. I know this is only in my personal experience, but I haven’t seen more than one or two “rejected” apps where I said, damn, it would’ve been awesome to have that. And frankly, the selection of apps is so large (150,000 and counting at this point) that almost any functionality you could ever need is there somewhere. Nor was I concerned with the multitasking, mainly because the rumor mill assures us it is coming in 4.0.

No, my main concern was simply whether the iPad served any useful purpose. We already have laptops and smartphones: is there any room left for a device like the iPad? My gut reaction was no,  the device, no matter how awesome it may be, was pretty much superfluous, and I had a feeling a lot of people would agree with me.

Turns out I may be wrong. The media blackout on iPad reviews lifted yesterday afternoon, and they’ve been pretty much all positive. I’ve read the reviews by Xeni Jardin at BoingBoing, Walt Mossberg at the WSJ, and David Pogue at the New York Times (my perennial favorite source for in-depth product reviews, Ars Technica won’t have theirs until next week, but I’m sure it’ll be worth the wait), and they’re all quite impressed. I got the sense from each of them that using the iPad is like getting at a scratch on your back: you didn’t even know it was there, but it feels so much better now that you’ve taken care of it.

Mossberg’s review in particular interested me: he said that in order to really test whether the iPad made sense as a “carry-everywhere” device, he stopped using his laptop altogether, did his work exclusively on his iPad, and tested whether he was still able to get everything done. The answer was, for the most part, yes. This makes the iPad very tempting to me: if it can do essentially everything a laptop can do, why carry around a laptop? The 12-hour battery life (I was quite skeptical of Apple’s 10-hour claim, but the reality is even better!) doesn’t hurt either. I’ll want to wait a while longer to see if the iPad really can do everything I would need it to (and you should too), but this is a pleasant surprise.

My current project in Natural Language Processing is to make a “language model” that is capable of generating random sentences. When I say “random” though, I don’t mean purely random- as in, the program picks words completely willy-nilly, which would produce gibberish like “helped spring to ferocious city while gaps thereby progression”. Instead, what the program does is scan some training data (in our case, hundreds and hundreds of newspaper articles), to build up a “model” of the English language: as it reads, it learns what words are more likely to come after other words. Then, when it’s all done, it generates sentences by first picking a word at random, then noting what words are likely to follow it and building up from there.

This is called an n-gram model, where n is the numer of words before the target word we take into consideration. For example, in a bigram model, we only consider the word immediately before the target word. So in that last sentence, our program would note, “if I see the word only, then consider is likely to come after it, and if I see the word immediately, then before is likely to come after it”. If, instead, we use a trigram model, then the program would note, “if I see immediately before then the next word is likely to be the“, and so on for any value of n we wish. Obviously higher values of n produce more accurate results, but it quickly leads to diminishing returns, and increasing n increases the amount of work the program has to do immensely.

You might think that a bigram model wouldn’t produce very good results, since it considers only one word at a time, but with our large training set, the sentences our program generated were surprisingly good. Obviously, in terms of content the sentences are gibberish, but it looks like almost-decent English, like the kind you would get if you ran a foreign website through Google Translate:

the first western partners will meet with the deal with an increase of the percentage of the united states .

bill clinton replied : it means that relations between the joint operations at the netherlands , including the need to comment on building socialism with no money supply natural gas off the map official sources said that , visited the korea has not less profitable .

local economy and expectations of the normal country are embarking on his speech at home , and the zulus must punish those who dare ignite a means to mount putuo .

it would give power-seekers pause for an effort in contempt of others who receive party organizations in japan .

Not bad at all, I think! Note in the third sentence, the phrase “… must punish those who dare ignite …” is used. That’s a six-word phrase that is legitimate English, and the program only had to look at one word at a time to produce it. Correct prepositional phrases, verb infinitives, and subject-predicate construction are all noticeable in the sample sentences too.

It got even better when we upped our program to use a trigram model. Take a look at these sample sentences:

some sources doubted the possibility of falling oil prices and help retire the agencys charges without admitting or denying any wrongdoing .

despite the problems of our nice little tricks on countervail and what-have-you , we are off of us to watch them grow here without some help .

Apart from the wonky “and help retire” bit in the middle, that first sentence is perfect English! It has subject-predicate agreement, and a relatively complex phrase, “without admitting or denying any wrongdoing”.

I also find it a little amusing that you can tell that our program was trained using newspaper articles: the sentences all have a very newspaper-like tone, and I’m not just talking about vocabulary.

The main point my professor is making in NLP this year is that researches who are trying to get computers to understand and generate English (or any other language) have all but abandoned a “rules-based approach”, where we try and teach the computer the rules of the language. Human languages are complicated things with many exceptions and a great deal of subjective interpretation, and trying to get a deterministic machine to emulate them has proved nigh-impossible with our current technology.

Instead, the wave of the future is “probability-based approaches”, where we feed the computer huge samples of text that humans have already made, and the computer builds a model of the language that way. Probability-based approaches are both easier and far more successful, as our results demonstrate: in just two weeks, three undergraduates with little to no linguistics knowledge got a computer to spit out legitimate English, something that rules-based approaches couldn’t do after decades.

As an aside note, this is one of the reasons Google is going to keep being successful in the years ahead. With access to vast quantities of data, they can teach a computer to do just about anything. Translation is a perfect example: pretty much everyone acknowledges that Google is way ahead of the competition in translating text between languages, because they can use their enormous datasets to do probability-based translation, whereas Babelfish et al. are stuck in the past with rules methods.