If you need a good reason to back change, it is that the current TLD system is heavily tilted towards English-speakers and a web culture in which linguistic meaning can be monotonically mapped to an alphabetic representation. But there are many languages which don't map unambiguously to a single style of romanization and several (such as Chinese) where the same alphabetic representation can imply many meanings.
This raises a few questions: isn't allowing users in non-English languages access to computing technology in a form suited to their own skills a good thing, and is there any hardwired technical reason why the TLD list can't be extended to become compatible with non-alphabetic encodings like Unicode?
The movies were thematically incongruous before the change, which is probably the reason Lucas made it. As far as I'm aware, there isn't a single conflict in the entire five films in which the aggressor wins.
http://www.stats.gov.cn/was40/detail?record=18&c ha nnelid=6697&presearchword=%B9%A4%D2%B5
You can decry the accuracy of statistical collection in China, but if you are going to offer a statement that flatly contradicts them, please reference. Official statistics are reasonable indicators of general trends, so how you turn a 17% industrial growth rate in 2003 negative is very unclear.
The only possible thing I can imagine is that you're only counting the public sector (state-owned enterprises). But claims of job loss are overstated here. Recent reforms are encouraging work units to restructure in ways which push workers off the books -- so the stats are misleading about the state of actual production.
Stop being shrill. Population growth clearly outpaced employment creation from November 2001-January 2004, exactly as Krugman says in his article. How you misread him as making a claim about the December-January period is beyond me, as is why you think any respectable economist would write a piece about unemployment TRENDS based on one month of statistical data.
But assuming you're are serious, I'd remind you that the recorded employment growth of 496,000 hardly outpaces growth in the size of the labour force 422,000. Nor does that slight gain really mean much when considered in historical context. At the very least, any suggestion that the US is in the midst of a job-creation euphoria driven by the June 2003 tax cuts, is sharply contradicted by a simple glance at the historical data:
The shortfall in Ontario has to do with quotas set by the provincial government on the numbers of doctors and nurses able to practice in the province, not any massive brain drain. It largely affects the ability of people to see family doctors, not receive emergency medical treatment.
If you want to make the case that health care in Canada is both more expensive and less efficient than it is in the US, you have an uphill road. Canadian health care may not look great when compared to many European models, but total health care spending as a percentage of GDP is still far lower in Canada than it is the US.
You can get monthly ADSL for about 100 kuai per month (1 kuai - 8.27 USD) if you're willing to commit for a year. Otherwise, a month of unlimited broadband in Beijing will set you back 120 kuai.
Access is slightly cheaper in the rural areas, and 5 gigs of prepaid wireless connectivity goes for 200 kuai per month through China Unicom.
So yes, you guys are being ripped off in the West. Move to China.
Stephen Hart would be more credible on film if he actually knew anything about it....
"Coruscant, the world-girdling capital city of Lucas' galactic republic, is a direct steal of Trantor, the planet-wide megalopolis in Isaac Asimov's "Foundation" novels"
If his complaint is of the "we've already had a film with a city/desert/water planet" sort, it's just silly. Moreover, it suggests Hart doesn't actually understand what Lucas is doing, such as why we continually get things like military forces being associated with water. If he is trying to argue that Lucas' sci-fi references are ad hoc.... he is just dead wrong. The primary film reference for Coruscant (there are several direct visual quotations) is to Fritz Lang's Metropolis, which predated Asimov's work by about two decades. Any sci-fi buff should pick right up on this, if only because the comparisons are so legion: Metropolis is also a story of father-son conflict with heavy themes of human mechanization, etc.
Hart fails to pick up on this or any of the other references (Blade Runner, Ben Hur, the Searchers, etc), but I'd suggest the problem lies in his ability to read films, not Lucas' ability to make them.
Simple: buy the same things (s)he is buying. As the US imports more Indian products, the value of the USD will fall against the Rupee. The price level will fall in the United States while rising in India. The end result will be a more expensive life for Indian programmers and a cheaper one for American ones.
Regardless, if you're pissed off about the current unemployment rate, there are more logical places to put your frustration than India, such as the Federal Reserve and the current administration. These groups have far more control over aggregate labor demand than anyone across the Pacific.
It runs on Windows, sucks at importing MS Word documents (silently removed all of the footnotes from one paper of mine), and installs an obnoxious screensaver whose only purpose seems announcing to the world what office productivity suite you run. Go Kingsoft!
But the Chinese language support is decent, and I'm going to guess that it saves its documents using the Guobiao standard -- which can be pretty useful if you want your documents to use the character-encoding standard mandated by the government.
What do you mean by double-byte enabled translation tools? We're talking character-conversion (jiantizi -> fantizi, etc.), not language translation, right?
Nice to read some posts by someone who obviously knows China.
That being said, I'd disagree that its the rural migrants are sopping up assembly-line jobs with the ODMs and that this is driving outsourcing to China. First of all, those guys go into construction, retail, and other kinds of jobs which don't require things like clean rooms. People in the US leap to the assumption that labour costs explain everything because of the dominance of the neoclassical economic paradigm, but realistically if this was just about labour there would be a hell of a lot more development in inland and northern China than there is now. Suzhou isn't exactly a haven for cheap wages the way... Fujian is, for instance.
I think its also a popular misconception that labour costs in China are significantly cheaper than elsewhere -- say parts of Indonesia, Malaysia. The difference is an easy investment channel from Taiwan (no language barrier), and ready access to the mainland market. China still has significant tariffs on imported laptops. The bulk of these might phase out over time in line with China's accession to the WTO, but I wouldn't count on them disappearing completely. If you want to sell your products in China there are still a lot of incentives to produce them there....
The events you describe were carried out by the Guomingtang (KMT) in 1947, not the Chinese Communist Party (CCP). The KMT was well-known for its visciousness against the local population, one of the reasons for the CCP victory. And incidentally, the fact that the KMT forced the local population to be educated in Mandarin is exactly the reason your attack on China is completely bogus.
As whoppers go, yours is about as big of one as mistaking France for Germany in WWI. But seriously, if you know so little about Chinese history you can't differentiate between two sides in one of the twentieth century's most important civil wars why are you posting anything at all regarding China, let alone these quasi-rascist diatribes which betray nothing but your ignorance of the country?
A deeper irony here is that if anyone was propping up the KMT during this period it was the United States, and exactly because of attitudes like yours: a reflexive and ideologically-driven misunderstanding of events half a world away.
(1) Why on earth should the Chinese speak English at their own press conferences?
(2) The language spoken in both Taipei and Beijing is putonghua, or standard mandarin. What language was this Taiwanese student supposed to be speaking?
(3) The phrase "capitalism with Chinese characteristics" was coined by Deng Xiaoping as part of the opening-up revolution that started with the 3rd Plenum of the 11th Party Congress in 1978. It refers to the expansion of the commercial sphere in China not its suppression.
You can buy a VCD player here in Beijing for $25 USD, and a regionless (*cough*) DVD/MP3/VCD combo for under $50. Since the average annual income is about $3000 USD, that's equivalent to someone making $24,000/year buying a $200 machine. Factor in VCD rentals at $0.20/day and DVD/VCDs on sale for between $1 and $1.50... its easy to see why owning this stuff is becoming pretty common.
Total population figure is irrelevant though. Even if people in rural Fujian aren't making enough money to buy a lot of DVDs, there are 16 million people in Beijing and several million more in the Yantze river delta. And when the population of just a few Chinese cities starts to rival countries like Germany... it makes a huge difference for international standards competition.
Dead right. Unless the inputed text really is sentence for sentence... you just end up with a bunch of garbage.
Makes me think the explosion of interest in this type of MT has less to do with its advantages over other systems than the limitations of current dictionaries. Who wants to spend years defining 300,000 words and phrases just on the off-chance it might be useful, right?
TI-99A: The VIAO is designed for extreme computing, driven by a double-capacity battery and equipped with an integrated CD-RW/DVD combo drive. It's arsenal includes a Memory Stick Media slot, for easy control of countless digital devices. It's body chassis is ultra-lightweight, and hardened against minor household accidents.
John: You'll find a way to destroy it.
TI-99A: Unlikely, I'm an obsolete design. The VIAO is a far more effective killing machine.
When I read economic "research" on questions like innovation in the software industry, I hardly see many people advancing neoclassical explanations for market development. I see more attention to path-dependent growth, standards competition, and things like critical mass. There are sound defensible models which - if right - suggest that patenting can be counterproductive in promoting innovation in certain industries.
So saying that these assumptions aren't off-base from an economics perspective doesn't say much. Why should we believe that your neoclassical assumptions accurately describe market dynamics?
Your post suggests that there are currently few incentives for research into basic algorithms. By your own assumptions however, this would lead me to ask why the United States currently has one of the richest legacies of innovation in software markets? It just doesn't make sense.
If anything, I'd agree with Paul above -- that commercial advances in algorithms occur in the commercial sector when necessary to solve concrete problems. In this case though, there is no direct link between the strength of patent protection and the incentive to innovate. Stronger patent protection would therefore be extremely unlikely to boost innovation, while doing an incredible amount to restrain market competition and innovation in other firms.
Incidentally, Xerox PARC may be a research lab, but I wasn't aware they conduct research into the sort of algorithms under discussion. Perhaps I'm wrong.
>> The only answer we have been able to come up with is patents, which are imperfect but an approximation of what we want.
This is a very market-oriented approach. Are you sure the assumptions you make (people won't innovate without financial incentives, and firms are the source of most innovation) are correct in this case?
It seems to me that most algorithms are not invented in the private sector, but come out of places like academia. After all, if it were otherwise, shouldn't there be a lot of firms out there that specialize in doing nothing BUT researching algorithms? With the exception of RSA (a commercialized academic discovery, not a commercial invention), I can't think of one off the top of my head.
An alternate solution to strengthening copyright is the public funding of basic applied research, or using government clout in the market (government procurement has historically driven many software sectors) to promote open standards.
If you have an open mind, why not read one of the better academic reviews of the new episodes?
You're probably right that Lucas isn't worried about losing his fan base though. I can't see anyone who really cares about commercial success making a series of films so intensely intellectual.
Re:I've always wondered about this
on
Swarm Intelligence
·
· Score: 2, Informative
I am an ever-struggling student of China, and am continually amazed by the quality of music I hear coming out of the Beijing rock scene. Beijing is without doubt one of the most vibrant places for cutting-edge rock and roll, perhaps because no one expects to get rich off CD sales. Even relatively old artists like Cui Jian are still producing great music.
Western record producers can gripe about piracy all they want, but it is simply a fact of life in China, and not just in music. A friend recently gave me a VCD of "Hero" - the new Zhang Yimou / Jet Li film. It is clearly a pirated copy, but is so visually stunning I plan to see it in theaters when I hit Beijing in two weeks (I don't know when it is scheduled to be released here....).
Realistically though, until someone explains to me why Chinese popular music is BETTER in quality and inventiveness than the stuff being played on MTV, I'll remain suspicious of arguments that tight copyright controls provide for better end-products.
p.s. Anyone hunting for good Chinese music should definitely check out Cui Jian. There was a really good documentary on China on PBS about a week ago that can be viewed here. It has a pretty decent soundtrack as well.
Slashdot Mirror
If you need a good reason to back change, it is that the current TLD system is heavily tilted towards English-speakers and a web culture in which linguistic meaning can be monotonically mapped to an alphabetic representation. But there are many languages which don't map unambiguously to a single style of romanization and several (such as Chinese) where the same alphabetic representation can imply many meanings.
This raises a few questions: isn't allowing users in non-English languages access to computing technology in a form suited to their own skills a good thing, and is there any hardwired technical reason why the TLD list can't be extended to become compatible with non-alphabetic encodings like Unicode?
The movies were thematically incongruous before the change, which is probably the reason Lucas made it. As far as I'm aware, there isn't a single conflict in the entire five films in which the aggressor wins.
The official government stats:
c ha nnelid=6697&presearchword=%B9%A4%D2%B5
http://www.stats.gov.cn/was40/detail?record=18&
You can decry the accuracy of statistical collection in China, but if you are going to offer a statement that flatly contradicts them, please reference. Official statistics are reasonable indicators of general trends, so how you turn a 17% industrial growth rate in 2003 negative is very unclear.
The only possible thing I can imagine is that you're only counting the public sector (state-owned enterprises). But claims of job loss are overstated here. Recent reforms are encouraging work units to restructure in ways which push workers off the books -- so the stats are misleading about the state of actual production.
Stop being shrill. Population growth clearly outpaced employment creation from November 2001-January 2004, exactly as Krugman says in his article. How you misread him as making a claim about the December-January period is beyond me, as is why you think any respectable economist would write a piece about unemployment TRENDS based on one month of statistical data.
But assuming you're are serious, I'd remind you that the recorded employment growth of 496,000 hardly outpaces growth in the size of the labour force 422,000. Nor does that slight gain really mean much when considered in historical context. At the very least, any suggestion that the US is in the midst of a job-creation euphoria driven by the June 2003 tax cuts, is sharply contradicted by a simple glance at the historical data:
http://www.bls.gov/webapps/legacy/cpsatab1.htm
The shortfall in Ontario has to do with quotas set by the provincial government on the numbers of doctors and nurses able to practice in the province, not any massive brain drain. It largely affects the ability of people to see family doctors, not receive emergency medical treatment.
If you want to make the case that health care in Canada is both more expensive and less efficient than it is in the US, you have an uphill road. Canadian health care may not look great when compared to many European models, but total health care spending as a percentage of GDP is still far lower in Canada than it is the US.
You can get monthly ADSL for about 100 kuai per month (1 kuai - 8.27 USD) if you're willing to commit for a year. Otherwise, a month of unlimited broadband in Beijing will set you back 120 kuai.
Access is slightly cheaper in the rural areas, and 5 gigs of prepaid wireless connectivity goes for 200 kuai per month through China Unicom.
So yes, you guys are being ripped off in the West. Move to China.
I was led to believe that commercial speech isn't constitutionally protected as free speech, which is probably the source of the distinction.
It is the same reason you can have limits on campaign contributions without imposing on the rights of "free speech" of corporations.
Hart fails to pick up on this or any of the other references (Blade Runner, Ben Hur, the Searchers, etc), but I'd suggest the problem lies in his ability to read films, not Lucas' ability to make them.
Simple: buy the same things (s)he is buying. As the US imports more Indian products, the value of the USD will fall against the Rupee. The price level will fall in the United States while rising in India. The end result will be a more expensive life for Indian programmers and a cheaper one for American ones.
Regardless, if you're pissed off about the current unemployment rate, there are more logical places to put your frustration than India, such as the Federal Reserve and the current administration. These groups have far more control over aggregate labor demand than anyone across the Pacific.
Yes they do. They just don't buy the services of American start-ups.
It runs on Windows, sucks at importing MS Word documents (silently removed all of the footnotes from one paper of mine), and installs an obnoxious screensaver whose only purpose seems announcing to the world what office productivity suite you run. Go Kingsoft!
But the Chinese language support is decent, and I'm going to guess that it saves its documents using the Guobiao standard -- which can be pretty useful if you want your documents to use the character-encoding standard mandated by the government.
What do you mean by double-byte enabled translation tools? We're talking character-conversion (jiantizi -> fantizi, etc.), not language translation, right?
Nice to read some posts by someone who obviously knows China.
That being said, I'd disagree that its the rural migrants are sopping up assembly-line jobs with the ODMs and that this is driving outsourcing to China. First of all, those guys go into construction, retail, and other kinds of jobs which don't require things like clean rooms. People in the US leap to the assumption that labour costs explain everything because of the dominance of the neoclassical economic paradigm, but realistically if this was just about labour there would be a hell of a lot more development in inland and northern China than there is now. Suzhou isn't exactly a haven for cheap wages the way... Fujian is, for instance.
I think its also a popular misconception that labour costs in China are significantly cheaper than elsewhere -- say parts of Indonesia, Malaysia. The difference is an easy investment channel from Taiwan (no language barrier), and ready access to the mainland market. China still has significant tariffs on imported laptops. The bulk of these might phase out over time in line with China's accession to the WTO, but I wouldn't count on them disappearing completely. If you want to sell your products in China there are still a lot of incentives to produce them there....
You're only digging yourself a deeper hole.
The events you describe were carried out by the Guomingtang (KMT) in 1947, not the Chinese Communist Party (CCP). The KMT was well-known for its visciousness against the local population, one of the reasons for the CCP victory. And incidentally, the fact that the KMT forced the local population to be educated in Mandarin is exactly the reason your attack on China is completely bogus.
As whoppers go, yours is about as big of one as mistaking France for Germany in WWI. But seriously, if you know so little about Chinese history you can't differentiate between two sides in one of the twentieth century's most important civil wars why are you posting anything at all regarding China, let alone these quasi-rascist diatribes which betray nothing but your ignorance of the country?
A deeper irony here is that if anyone was propping up the KMT during this period it was the United States, and exactly because of attitudes like yours: a reflexive and ideologically-driven misunderstanding of events half a world away.
(1) Why on earth should the Chinese speak English at their own press conferences?
(2) The language spoken in both Taipei and Beijing is putonghua, or standard mandarin. What language was this Taiwanese student supposed to be speaking?
(3) The phrase "capitalism with Chinese characteristics" was coined by Deng Xiaoping as part of the opening-up revolution that started with the 3rd Plenum of the 11th Party Congress in 1978. It refers to the expansion of the commercial sphere in China not its suppression.
A troll is a troll is a troll....
You can buy a VCD player here in Beijing for $25 USD, and a regionless (*cough*) DVD/MP3/VCD combo for under $50. Since the average annual income is about $3000 USD, that's equivalent to someone making $24,000/year buying a $200 machine. Factor in VCD rentals at $0.20/day and DVD/VCDs on sale for between $1 and $1.50... its easy to see why owning this stuff is becoming pretty common.
Total population figure is irrelevant though. Even if people in rural Fujian aren't making enough money to buy a lot of DVDs, there are 16 million people in Beijing and several million more in the Yantze river delta. And when the population of just a few Chinese cities starts to rival countries like Germany... it makes a huge difference for international standards competition.
Dead right. Unless the inputed text really is sentence for sentence... you just end up with a bunch of garbage.
Makes me think the explosion of interest in this type of MT has less to do with its advantages over other systems than the limitations of current dictionaries. Who wants to spend years defining 300,000 words and phrases just on the off-chance it might be useful, right?
I guess it's just a question of mailing lists then. Here's to hoping I stay off whatever you've managed to get put on.... ;)
The vast majority of spam hitting *my* inbox starts coming in at around 9am California time, and then peters out early evening on the East Coast.
What's this about an international problem again?
TI-99A:
The VIAO is designed for extreme computing, driven by a double-capacity battery and equipped with an integrated CD-RW/DVD combo drive. It's arsenal includes a Memory Stick Media slot, for easy control of countless digital devices. It's body chassis is ultra-lightweight, and hardened against minor household accidents.
John:
You'll find a way to destroy it.
TI-99A:
Unlikely, I'm an obsolete design. The VIAO is a far more effective killing machine.
When I read economic "research" on questions like innovation in the software industry, I hardly see many people advancing neoclassical explanations for market development. I see more attention to path-dependent growth, standards competition, and things like critical mass. There are sound defensible models which - if right - suggest that patenting can be counterproductive in promoting innovation in certain industries.
So saying that these assumptions aren't off-base from an economics perspective doesn't say much. Why should we believe that your neoclassical assumptions accurately describe market dynamics?
Your post suggests that there are currently few incentives for research into basic algorithms. By your own assumptions however, this would lead me to ask why the United States currently has one of the richest legacies of innovation in software markets? It just doesn't make sense.
If anything, I'd agree with Paul above -- that commercial advances in algorithms occur in the commercial sector when necessary to solve concrete problems. In this case though, there is no direct link between the strength of patent protection and the incentive to innovate. Stronger patent protection would therefore be extremely unlikely to boost innovation, while doing an incredible amount to restrain market competition and innovation in other firms.
Incidentally, Xerox PARC may be a research lab, but I wasn't aware they conduct research into the sort of algorithms under discussion. Perhaps I'm wrong.
>> The only answer we have been able to come up with is patents, which are imperfect but an approximation of what we want.
This is a very market-oriented approach. Are you sure the assumptions you make (people won't innovate without financial incentives, and firms are the source of most innovation) are correct in this case?
It seems to me that most algorithms are not invented in the private sector, but come out of places like academia. After all, if it were otherwise, shouldn't there be a lot of firms out there that specialize in doing nothing BUT researching algorithms? With the exception of RSA (a commercialized academic discovery, not a commercial invention), I can't think of one off the top of my head.
An alternate solution to strengthening copyright is the public funding of basic applied research, or using government clout in the market (government procurement has historically driven many software sectors) to promote open standards.
You're probably right that Lucas isn't worried about losing his fan base though. I can't see anyone who really cares about commercial success making a series of films so intensely intellectual.
Try http://www-inst.eecs.berkeley.edu/~pshyu/engnews/e ngnews.html
I am an ever-struggling student of China, and am continually amazed by the quality of music I hear coming out of the Beijing rock scene. Beijing is without doubt one of the most vibrant places for cutting-edge rock and roll, perhaps because no one expects to get rich off CD sales. Even relatively old artists like Cui Jian are still producing great music.
Western record producers can gripe about piracy all they want, but it is simply a fact of life in China, and not just in music. A friend recently gave me a VCD of "Hero" - the new Zhang Yimou / Jet Li film. It is clearly a pirated copy, but is so visually stunning I plan to see it in theaters when I hit Beijing in two weeks (I don't know when it is scheduled to be released here....).
Realistically though, until someone explains to me why Chinese popular music is BETTER in quality and inventiveness than the stuff being played on MTV, I'll remain suspicious of arguments that tight copyright controls provide for better end-products.
p.s. Anyone hunting for good Chinese music should definitely check out Cui Jian. There was a really good documentary on China on PBS about a week ago that can be viewed here. It has a pretty decent soundtrack as well.