But if only I have a file and want to distribute it to those 10 people in my office, using BitTorrent is much faster than copying it to them one after another -- precisely because of the bandwidth throttle on my end.
First of all, "Asian scripts" is a totally bogus term: East Asian scripts (Chinese and derivatives, aka CJK), which are logographic, has no relation whatsoever with other Asian scripts (e. g. Mongolian, Thai, Indic, Arabic etc.), which are alphabetic and very much related to non-Asian alphabetic scripts (e. g. Greek and derivatives like Latin).
Second of all, neither the CJK scripts nor the other Asian scripts has a stronger emphasis on line thickness than non-Asian scripts. Including line thickness as an additional parameter would certainly improve OCR for CJK, but no more than it would for any other script.
From TFA: "Citing sources in the defense contracting and intelligence consulting community, the iDefense report unambiguously declares that the Chinese government was, in fact, behind the effort."
Right, for what possible sinister reason could people in the American "defense contracting and intelligence consulting community" have to paint China as a threat to US national security?
BTW, in day-to-day operations, German's ICE and Japan's Sinkansen often go beyond 300 km/h. Frace's TGV never does, and Canada's Bombaardier doesn't even work well above 200 km/h. The expected top speed in day-to-day operations for this route is about 350 km/h with the German trains and 300 km/h with the Japanese.
But as car-lovers know, more difficult than driving fast is braking fast. The new Chinese trains uses Siemens' braking technology that provides five (!) entirely different, independently operating braking systems, of which only one needs to work (that's 400% redundant, obsessive even for German standards) in order to completely stop the train within a few km. That's why this route is not only going to be one of the fastest in the world, but also one of the most densely operated (i. e. shortest safety distance between one train and the next).
Someone in my family works for Siemens as a senior member of the China High-Speed Rail project (not to be confused with the China Maglev project, for which Siemens is also a partner). We've talked about it quite often - and fairly extensively yesterday. Here are a few details:
The technologies of all four major high-speed rail system in the world - Germany's ICE, Japan's Sinkansen, France's TGV and Canada's Bombardier (in order of overall technological advancement) - have come together in China, though rather reluctantly. When the Chinese started the project years ago, they did something very clever: Instead of picking one of the four systems (which is what people normally do), they gave all four a pilot contract each. The one showing the best result in its pilot would then be chosen as the main partner, they said, making all four competing like crazy - routinely investing more resources than they've originally planed. The Chinese are not concerned about significant waste due to incompatibility between the pilot products, since all four are building to the specs written by the Chinese.
Now, years later, the Canadians and the French are practically washed out, even though some of their technologies have contributed to the new Chinese system. The Germans and the Japanese remain - as initially expected - the main competitors - or, reluctant partners for the Chinese. The vast majority of heavy lifting on the technological front is done by the Germans (which was also expected, since even the Japanese system was originally based on German designs), but the Japanese have the advantage that their pilot has started earlier (the Chinese intentionally delayed the German pilot in order to ransom a below-value price).
The record speed, for example, was achieved using two joined trains - of four sections each - built by Siemens in Germany and put together in China. Those are the only two German trains current available for this route. All the other trains are Japanese, and they're what people see on most new footages. But the top speed the Japanese trains (on the same route) can reach are significantly lower - about 350 km/h, or >10% less than the German record. Plus, while the German rains got to 395 km/h in standard configuration - with two tracking (active) and two tracked (passive) sections in each train - the Japanese had to cheat - using three tracking and only one tracked section in each train - in order to reach their 350 km/h.
As someone has mentioned above, there exist a TGV speed record that's much higher still, but that's a record nobody in the industry takes seriously, because it was achieved with a totally crazy, not nearly practical configuration of train sections. It's a fake number, period.
The bottom line is, for the original cost of one project, China has managed to get more than twice the amount worth of know-how (all legally via proper technology transfer contracts), and is now itself among the leading players of the industry. For the upcoming US high-speed rail system, the Chinese has offered a bid with a price tag 1/3 lower than anybody else...
... isn't software v. hardware, but speed v. functionality, i. e., in the history of most software, the decrease in speed is disproportional to the increase in functionality. Of course, "disproportional" is subjective, and new, advanced functionalities are generally more complicated and resource intensive than old, basic ones. So a simple reverse-linear relationship might be unrealistic, but when many software don't even manage to beat the reverse-quadratic ratio, there's definitely something wrong.
Microsoft and Google should just publish the exact data exchange protocol used by their respective "smart search" features -- and keeps those documentations up to date, of course.
The protocols are gonna get out, anyway -- someone will snoop them out soon enough. Better have an official documentation than endless wild rumors, and the whole thing would hardly cost any resources.
This chip might be indeed unclonable, i. e. it cannot be emulated by another chip with the same look and same inner structure. But it's still very much copyable, i. e. it can be emulated by another chip with a different look and/or inner structure.
Think of a mechanical door key made of platinum: A normal locksmith may not be able to clone it since he doesn't have platinum key blanks and his tools are not tuned to process platinum. But he can still copy the key using a usual key blank, and this copy, though not made of platinum and thus not a clone, can still emulate the original key, i. e. opening the respective door.
Since the whole purpose of RFID is contactless access, the chips are usually out of sight during real world usage. That means clonability is nearly irrelevant, emulatability is the real issue.
This thing isn't new at all: Eighteen years ago, at the 1990 Asian Games in Beijing, the organizers already managed to control the rain quite successfully. For instance, the opening ceremony which would have been disrupted by rain without intervention, ran smoothly in fairly sunny weather instead.
The technique is simple: Detect in advance the clouds which could cause rain in Beijing, then send airplanes to spread special dust particles to cause those clouds to rain immediately, thus "empty" them before they reach Beijing.
I'm quite convinced the Chinese aren't the only ones who's done this.
A lot of what Eckel is saying is basically "Python should be more like other languages" - not because they're better, but because I'm more used to them. Obviously that's totally ridiculous, yet not surprising: If you look at his resume, it seems he's far more familiar with other languages than Python. I seriously doubt his credential - let alone objectiveness - to question Python's design.
The Shijinshan Amusement Park was built more than 20 years ago. Not sure if the PRC has established diplomatic ties with USA yet, yet alone intellectual property treaties.
Also, twenty years ago there was practically not such thing as "intellectual property" in China (ah, the good old days). There was a very famous brand of candy from Shanghai named "Mickymouse", with Disney's Mickymouse as its logo. Later, when China opened its market to the West, Disney realized that in China, the "Mickymouse" trademark was legally owned by the candy factory, so it paid big money and brought it back. It was reported in the newspaper that many Chinese children cried that day as their beloved "Mickymouse" candy was no more (oh, the evil Americans)...
When I finally worked it all out and got arccos(sqrt(3)/sqrt(5)) for (iii) I decided that there perhaps wasn't a simple solution for (ii)
There indeed is a very simple solution for (ii): The angle in question is equal to the angle A1EC1, which is a rectangle. To prove that, simply calculate A1E, C1E and show A1C1^2 = A1E^2 + C1E^2.
No vector calculation is necessary for any part of the test.
Yes, I came to the same result as you using basic geometry, without any vector calculation - see my comment above, #18876849.
Vector calculation is comparatively advanced stuff, I believe nowhere in the world it is taught in the first nine years of school - which is the legal minimal of school education in most countries.
The Chinese test does indeed require more knowledge to answer, especially part (iii). Here is the (probably most direct) solution:
First we establish a point F on AC so AD || BF. It can be easily shown that the angel between AC and BC1 is equal to the angel FBC1.
Further it's not hard to calculate that BC = DC = sqrt(12), BF = DF = sqrt(3), AC = 4, AE = AF = 1, so FC = 2.
In the rectangle triangle BCC1 where BC = sqrt(12) and CC1 = sqrt(3), BC1 = sqrt(15). In the rectangle triangle FCC1 where FC = 2 and CC1 = sqrt(3), FC1 = sqrt(7).
In the triangle FBC1 where BF = 2, BC1 = sqrt(15) and FC1 = sqrt(7), use the following equation (generally applicable to any triangle) to calculate the angle FBC1: AC^2 = AB^2 + BC^2 - 2*AB*AC*cos(ABC)
The angle in question is roughly 39.23 degree.
Now the triagle FBC1 is not a "common" triangle, and a angle of 39.23 degree is not a "common angle". The equation above is not something normal kids learn in grade school - as opposed to the Pythagorean theorem, which is all you need to answer the English test.
Ballmer: Our great leader, his Billness, demonstrates his heroics to the world by defeating four millions of spam every day!
Advisor:... (inaudible)
Ballmer: (whispering) What? Bug reports don't count as spam? What about customer complaints, they sure do, don't they? Well, I thought all that stuff we get and then move to the trash box without reading is spam... No? Fsck! So how much does the Boss gets?
Advisor:... (still inaudible)
Ballmer: (whispering angrily) Fsck you! How am I suppose to do propaganda with those kind of numbers?! Wait, I have an ingenious idea! Tell this Hotty Mail or whatever we own, let them loosen their anti-spam policy some more, and cc all spam sent from there to the Boss!
Advisor:... (leaving hastily)
Ballmer: Our great leader, his Billness, demonstrates his heroics to the world by defeating four millions of spam every year!
It'd be childish to blindly assume criminals were cleverer than vendors. They're not. Instead, they:
- have more efficient "cost structure" and thus more manpower, and
- are bound to less limitations (practically none, to be exakt - other than the law of physics, that is, if you must count that).
Slashdot Mirror
... so we better get some pollution going.
But if only I have a file and want to distribute it to those 10 people in my office, using BitTorrent is much faster than copying it to them one after another -- precisely because of the bandwidth throttle on my end.
Right, 'cause it's much nicer to do business in countries where corporate can push government around (like in the US of A).
It's one thing to say there is insufficient evidence, but *no* evidence?!
First of all, "Asian scripts" is a totally bogus term: East Asian scripts (Chinese and derivatives, aka CJK), which are logographic, has no relation whatsoever with other Asian scripts (e. g. Mongolian, Thai, Indic, Arabic etc.), which are alphabetic and very much related to non-Asian alphabetic scripts (e. g. Greek and derivatives like Latin).
Second of all, neither the CJK scripts nor the other Asian scripts has a stronger emphasis on line thickness than non-Asian scripts. Including line thickness as an additional parameter would certainly improve OCR for CJK, but no more than it would for any other script.
From TFA: "Citing sources in the defense contracting and intelligence consulting community, the iDefense report unambiguously declares that the Chinese government was, in fact, behind the effort."
Right, for what possible sinister reason could people in the American "defense contracting and intelligence consulting community" have to paint China as a threat to US national security?
I was talking about this new Wuhan-Guangdong route, and you're citing numbers for the entire Shinkansen network.
BTW, in day-to-day operations, German's ICE and Japan's Sinkansen often go beyond 300 km/h. Frace's TGV never does, and Canada's Bombaardier doesn't even work well above 200 km/h. The expected top speed in day-to-day operations for this route is about 350 km/h with the German trains and 300 km/h with the Japanese.
But as car-lovers know, more difficult than driving fast is braking fast. The new Chinese trains uses Siemens' braking technology that provides five (!) entirely different, independently operating braking systems, of which only one needs to work (that's 400% redundant, obsessive even for German standards) in order to completely stop the train within a few km. That's why this route is not only going to be one of the fastest in the world, but also one of the most densely operated (i. e. shortest safety distance between one train and the next).
Someone in my family works for Siemens as a senior member of the China High-Speed Rail project (not to be confused with the China Maglev project, for which Siemens is also a partner). We've talked about it quite often - and fairly extensively yesterday. Here are a few details:
The technologies of all four major high-speed rail system in the world - Germany's ICE, Japan's Sinkansen, France's TGV and Canada's Bombardier (in order of overall technological advancement) - have come together in China, though rather reluctantly. When the Chinese started the project years ago, they did something very clever: Instead of picking one of the four systems (which is what people normally do), they gave all four a pilot contract each. The one showing the best result in its pilot would then be chosen as the main partner, they said, making all four competing like crazy - routinely investing more resources than they've originally planed. The Chinese are not concerned about significant waste due to incompatibility between the pilot products, since all four are building to the specs written by the Chinese.
Now, years later, the Canadians and the French are practically washed out, even though some of their technologies have contributed to the new Chinese system. The Germans and the Japanese remain - as initially expected - the main competitors - or, reluctant partners for the Chinese. The vast majority of heavy lifting on the technological front is done by the Germans (which was also expected, since even the Japanese system was originally based on German designs), but the Japanese have the advantage that their pilot has started earlier (the Chinese intentionally delayed the German pilot in order to ransom a below-value price).
The record speed, for example, was achieved using two joined trains - of four sections each - built by Siemens in Germany and put together in China. Those are the only two German trains current available for this route. All the other trains are Japanese, and they're what people see on most new footages. But the top speed the Japanese trains (on the same route) can reach are significantly lower - about 350 km/h, or >10% less than the German record. Plus, while the German rains got to 395 km/h in standard configuration - with two tracking (active) and two tracked (passive) sections in each train - the Japanese had to cheat - using three tracking and only one tracked section in each train - in order to reach their 350 km/h.
As someone has mentioned above, there exist a TGV speed record that's much higher still, but that's a record nobody in the industry takes seriously, because it was achieved with a totally crazy, not nearly practical configuration of train sections. It's a fake number, period.
The bottom line is, for the original cost of one project, China has managed to get more than twice the amount worth of know-how (all legally via proper technology transfer contracts), and is now itself among the leading players of the industry. For the upcoming US high-speed rail system, the Chinese has offered a bid with a price tag 1/3 lower than anybody else...
From TFA: "I suppose that adds up, 90 per cent less code equates to ten times faster."
Really? I suppose you can write haiku ten times faster than stream-of-conscious recordings, too?
There's one thing that computer code and natural language text have in common: For both, confusing "writing" with "typing" is moronic.
... isn't software v. hardware, but speed v. functionality, i. e., in the history of most software, the decrease in speed is disproportional to the increase in functionality. Of course, "disproportional" is subjective, and new, advanced functionalities are generally more complicated and resource intensive than old, basic ones. So a simple reverse-linear relationship might be unrealistic, but when many software don't even manage to beat the reverse-quadratic ratio, there's definitely something wrong.
BTW, the author of TFA is the submitter of this "story" (email address matches byline).
Microsoft and Google should just publish the exact data exchange protocol used by their respective "smart search" features -- and keeps those documentations up to date, of course.
The protocols are gonna get out, anyway -- someone will snoop them out soon enough. Better have an official documentation than endless wild rumors, and the whole thing would hardly cost any resources.
This chip might be indeed unclonable, i. e. it cannot be emulated by another chip with the same look and same inner structure. But it's still very much copyable, i. e. it can be emulated by another chip with a different look and/or inner structure.
Think of a mechanical door key made of platinum: A normal locksmith may not be able to clone it since he doesn't have platinum key blanks and his tools are not tuned to process platinum. But he can still copy the key using a usual key blank, and this copy, though not made of platinum and thus not a clone, can still emulate the original key, i. e. opening the respective door.
Since the whole purpose of RFID is contactless access, the chips are usually out of sight during real world usage. That means clonability is nearly irrelevant, emulatability is the real issue.
With this kind of excessive use of random data, isn't hash collision to be expected pretty soon?
This thing isn't new at all: Eighteen years ago, at the 1990 Asian Games in Beijing, the organizers already managed to control the rain quite successfully. For instance, the opening ceremony which would have been disrupted by rain without intervention, ran smoothly in fairly sunny weather instead.
The technique is simple: Detect in advance the clouds which could cause rain in Beijing, then send airplanes to spread special dust particles to cause those clouds to rain immediately, thus "empty" them before they reach Beijing.
I'm quite convinced the Chinese aren't the only ones who's done this.
A lot of what Eckel is saying is basically "Python should be more like other languages" - not because they're better, but because I'm more used to them. Obviously that's totally ridiculous, yet not surprising: If you look at his resume, it seems he's far more familiar with other languages than Python. I seriously doubt his credential - let alone objectiveness - to question Python's design.
The Shijinshan Amusement Park was built more than 20 years ago. Not sure if the PRC has established diplomatic ties with USA yet, yet alone intellectual property treaties. Also, twenty years ago there was practically not such thing as "intellectual property" in China (ah, the good old days). There was a very famous brand of candy from Shanghai named "Mickymouse", with Disney's Mickymouse as its logo. Later, when China opened its market to the West, Disney realized that in China, the "Mickymouse" trademark was legally owned by the candy factory, so it paid big money and brought it back. It was reported in the newspaper that many Chinese children cried that day as their beloved "Mickymouse" candy was no more (oh, the evil Americans)...
There indeed is a very simple solution for (ii): The angle in question is equal to the angle A1EC1, which is a rectangle. To prove that, simply calculate A1E, C1E and show A1C1^2 = A1E^2 + C1E^2.
No vector calculation is necessary for any part of the test.
Yes, I came to the same result as you using basic geometry, without any vector calculation - see my comment above, #18876849.
Vector calculation is comparatively advanced stuff, I believe nowhere in the world it is taught in the first nine years of school - which is the legal minimal of school education in most countries.
Minor corrections:
By "BF = DF = sqrt(3)" I meant "BE = DE = sqrt(3)". BTW, BF = DF = BA = DA = 2.
In latter "AC" in the equation is a typo for "BC", so the correct form is:
AC^2 = AB^2 + BC^2 - 2*AB*BC*cos(ABC)
The Chinese test does indeed require more knowledge to answer, especially part (iii). Here is the (probably most direct) solution:
First we establish a point F on AC so AD || BF. It can be easily shown that the angel between AC and BC1 is equal to the angel FBC1.
Further it's not hard to calculate that BC = DC = sqrt(12), BF = DF = sqrt(3), AC = 4, AE = AF = 1, so FC = 2.
In the rectangle triangle BCC1 where BC = sqrt(12) and CC1 = sqrt(3), BC1 = sqrt(15).
In the rectangle triangle FCC1 where FC = 2 and CC1 = sqrt(3), FC1 = sqrt(7).
In the triangle FBC1 where BF = 2, BC1 = sqrt(15) and FC1 = sqrt(7), use the following equation (generally applicable to any triangle) to calculate the angle FBC1:
AC^2 = AB^2 + BC^2 - 2*AB*AC*cos(ABC)
The angle in question is roughly 39.23 degree.
Now the triagle FBC1 is not a "common" triangle, and a angle of 39.23 degree is not a "common angle". The equation above is not something normal kids learn in grade school - as opposed to the Pythagorean theorem, which is all you need to answer the English test.
Ballmer: Our great leader, his Billness, demonstrates his heroics to the world by defeating four millions of spam every day!
Advisor: ... (inaudible)
Ballmer: (whispering) What? Bug reports don't count as spam? What about customer complaints, they sure do, don't they? Well, I thought all that stuff we get and then move to the trash box without reading is spam... No? Fsck! So how much does the Boss gets?
Advisor: ... (still inaudible)
Ballmer: (whispering angrily) Fsck you! How am I suppose to do propaganda with those kind of numbers?! Wait, I have an ingenious idea! Tell this Hotty Mail or whatever we own, let them loosen their anti-spam policy some more, and cc all spam sent from there to the Boss!
Advisor: ... (leaving hastily)
Ballmer: Our great leader, his Billness, demonstrates his heroics to the world by defeating four millions of spam every year!
It'd be childish to blindly assume criminals were cleverer than vendors. They're not. Instead, they: - have more efficient "cost structure" and thus more manpower, and - are bound to less limitations (practically none, to be exakt - other than the law of physics, that is, if you must count that).
I think this is deliberate, to test how a client deals with broken code.