The US doesn't have "control of the internet", so it cannot be relinquishing what it doesn't have. The ICANN being US-based doesn't give much real control over IP packets travelling on some fiber halfway around the world from DC. Even if ICANN was a government agency it wouldn't. It just allows to vaguely arbitrate over domain names and IP number disputes that have relatively faint commercial implications. And even then the US feds would have to use indirect influence on ICANN.
I agree with the article that while India and China have been twin rising stars in the tales of ideological globalists like NYT's Thomas Friedman, there is a huge difference between the two : while China will be a superpower by the end of this century, at that time India will still be a third world country by far. With its caste structure, its irremediable lack of infrastructure and ressources to support its population, its relative submissiveness to western political pressure, the tendency of its educated elite to go live and work abroad the second they have a chance to, the best that can happen to India in the mid term is to nurture a developped sub-economy that will give it the global importance of, say, Italy, the UK or France.
Okay so that means the end of OS X within two years which brings a couple of benefits: - Apple execs shouldn't be afraid to run out of cat names - good for Linux since OS X was so far the only Unix embedded in a single, stable, nice GUI with drivers for all the gadgets in the market, it tended to divert Linux people who want their shell, their X11, their clean POSIX and their native Emacs/make.
They don't absolutely need these planes to be V/STOL since they are to be used on their future, full-size carrier which, unlike their existing mini-carriers will be able to operate any kind of naval jet like F/A-18 or Rafale M provided they are fitted with a catapult.
The design of their future carriers is already supposed to be largely shared with that of France's second carrier which is supposed anyway to operate Rafales. So the decision to dump V/STOL would simplify things actually.
I don't think that the UK has ordered any JSF yet. What they did is pay part of the development costs, in return for the promise of a share of the industrial pie when the plane enters production. Their high level of financial participation was also supposed to buy them some input in the specification and some sharing of classified technologies, but the Americans largely didn't carry out that part of the deal, which has provoked transatlantic tension lately.
The JSF is supposed to equip the RNs future carrier ships around around 2015. However as a response to the US Congress looking at cancelation of plans for a Rolls-Royce engine equiped version of the JSF, the British have hinted that they could very well start developing a naval version of the Eurofighter Typhoon, or even consider the already operationnal naval version of the French Rafale.
I'm afraid this person who I assume is a fellow French national, is a little bit misinformed himself about the workings of his own country : the Cour de Cassation IS the highest court for civil law. There are three institutions that cover what the Supreme Court does in the US : The Cour de Cassation and the Council of State are the highest appeals for respectively civil law and administrative law. The former rules whether justice was administered properly in cases involving crimes and felonies of people or businesses. The later judges whether the State and the local governments act within their legit powers and is the highest appeal for people who, to put in in American terms, "sue the government". And finally the Constitutionnal Council censors bills when they are incompatible with the constitution or international law, _before thay are made into law_. It is not possible to appeal to it once the law is signed, unlike with the Supreme Court.
Yes : to break the ruling of a lower court. That's what the cour de cassation, which is the highest court for civil law, does. Note that the CdC does not rule on the content of cases, but only on the way justice was administered.
I don't think it's useless in the real world : first, complete formal validation of software is used in the real world for critical systems, even though it's considered anti-economical to use it for general programming. However it's more a matter of attitudes than of fundamental methodologies. And attitudes are shifting. I work at a semi-conductor company in the domain of formal hardware validation, and researchers here have noticed a significant academic brain drain from h/w validation towards formal software validation since Microsoft put its chips in it.
The Microsoft example is significant when I speak of "attitudes". The point is not to have every programmer mathematicaly prove properties on each and every function they write in an application (although core libs and OSs could be subject to that). The point is to have people with a _perspective_ of understanding what they write rather than an attitude of cutting-and-pasting "patterns" or algorithms from textbooks. What MS has stated is that they were doing this attitude shift.
If I had to interview people for a developper position, here's what I would do : I would ask them to write a "sort" function, preferably in whatever language they would be assigned to work in. And then I would review their code with them, not really paying any attention to whether the thing actually works or compiles or not. What I would ask them is "tell me why your code works". From there:
- VERY BAD ANSWER : "It works because look, I wrote a quicksort" - QUITE BAD ANSWER ": "It works because it compiles and look, I give it this particular list and lo! I get a sorted list back." - SOMEWHAT REASSURING ANSWER : "It works because I wrote this test unit and I analyzed the range of possible inputs and generated this stimulation set that covers all the border situations". - WHEN ARE YOU AVAILABLE TO START ? : "It works because what I want is for the resulting list to contain all the entries in the input list, and only these, and that I want each element of the resulting list to be less or equal to its successor. When I call function qsort, as you can see here this particular line in my code means that the result will have its pivot in the right position, so by applying this invariant recursively, etc..."
I think that many people who are programming today, and even leaving code long term deep into production systems, should be kicked out of the field, when they don't really know what they're doing and have had a two years training of "programming-is-a-bit-like-spelling-out-a-cooking- recipe" style Java drill.
Programming requires an intellectual discipline, and an understanding of abstract mathematical concepts that is just not anybody's. Then we'll see systems working much better.
When I was a kid in the 70s, my parents were very enthusiastic in impregnating me with the technology that would later make me successful in life. They have really inspired my life and made my calling very clear early on in my life.
However right now I'm out of a job. Anyone needs someone who's REALLY GOOD at punching COBOL programs on cards ?
I don't know how intentionnal it is, I assume it is intentionnal : in computer science a Y combinator is, basically, a function that takes a function as an argument and returns a fixpoint to the function. so for any function f, f(Y(f))=Y(f). Oh, what's the hell : just check out the wikipedia.
My first computer was an EXL-100, from Exelvision of France, in 6th grade I believe which was, like, 1985. The company was started I believe by one or more French TI engineers who built their stuff from a TI chipset (rather than a single uproc). Its remarkable features were a "pro" form factor with a wireless (IR) keyboard separate from the CPU and some sound DAC used as a "voice synthetizer" giving pretty amazing results for the era.
One of a flurry of simple home computers from European companies in the early 80s, it didn't benefit from the government support (in the form of education sales) of its compatriots TO-7 and MO-5 from state-owned company Thomson. It didn't have the "cool games" and one year later I switched to a CPC-464 from Amstrad of the UK.
France does not have a pure civil law system, just as so-called "common law" countries don't have a pure precedent-based system. After all one of the sources of US law is a written source : the constitution.
In theory, the only law applicable in France is written by the legislative branch. The judiciary is supposed to only work as a machine, applying the following pseudo-code:
foreach (case) {
check out facts
look up relevant article of law
pronounce judgement }
Precedent does exist, but is viewed more as a labor saver, a little bit like speeding up an algorithm by caching intermediate results that can be used more than once.
Most of the written law in France is arguably well designed, some of it, most famously Napoleon's civil code, having been barely touched in more than 200 years.
The benefit of a written law system then is that in many cases, if you want to know whether you're right to do something, you just look it up in the book (or on www.legifrance.gouv.fr), instead of giving $1000 to a lawyer for advice and crossing your fingers.
However some other codes are not working so well, two examples would be the intellectual property code, which is faced with a technology shift of millenary proportions, and the labor code, which is creaking under thousands of categorical-interest driven articles that are largely contradictory and contrarian to economic common sense.
In these domains, the precedents take center stage.
When I was a teenager in the '80s in France my parents rented a vacation home for two or three weeks, and as usual the owners had left a bunch of old books and magazines to garnish the bookshelves. Among them, tourism and travel guides to the USSR, which led my parents to frown upon having rented the place from people with communist sympathies. Browsing this guides was fascinating though, and I vividly remember how they "diplomatically" were introducing the potential traveller to the photography restrictions. Reading this, the notion of the authorities of a country restricting what could be photographed in it, particarly outdoors, sounded bizarre and impracticable...
So I don't get the logic behind the politically-correct Intel bashing. On the one hand, one hears that Intel is bad because they have carried on with binary compatibility since the microprocessor pre-history (the 8086, even the 8080 to a certain extent), creating an architecture that is indeed in some places a bit quaint.
Then on the other hand we are supposed to believe that AMD is genius for having led the way in moving to 64 bits through extension of that much-reviled x86 architecture rather than by starting from scratch with an innovative and modern architecture such as EPIC.
So the bottom line is the Intel is always wrong, right ?
Oh anyway, go and count how many www.top500.org supercomputers in the world are Intel powered, and how many are AMD powered. People who build supercomputing clusters must be clueless joe's who buy into Intel's evil marketing messages, right ? Because I see more "Itanics" than Opterons there, and way more Xeons than either of these.
"Military Police" is a very bad translation for Gendarmerie. The Gendarmerie is pretty much just a regular police force. Police in France is almost exclusively a national-level organization, unlike for example in the US where it is locally organized (city, county, state). Police is divided in two categories : the National Police, made of civil servants under the interior ministry, and the Gendarmerie, which is a branch of the military. Both do essentially the same job, wear similar uniforms, and are assigned to different districts, with the Police dealing with urban areas and the Gendarmerie with rural areas.
Differences are that the Gendarmes almost never work in plain clothes. They occasionaly wear olive green fatigues for field work away from civilian contact. They can be sent overseas as easily as the other military people, which is very convenient for France in providing trained civilian-style law enforcement capabilities in international peace keeping missions.
Other countries have or have had similar forces : the Gendarmerie in Belgium, the Guardia Civil in Spain, both of which I believe have been "civilianized". The Carabinieri in Italy.
It's hasty translation. "Gouvernement" in France means "Prime Minister's cabinet", which, along with the presidency, form what Americans would call the executive branch of government.
The Prime Minister and his cabinet are always according to the majority elected to the lower house of parliament, which is not the case of the president who is elected independently.
However the MPs (députés), even though they theoretically can draw up bills or reject the gouvernement's proposed ones, usually acts as a rubber stamp to the cabinet designed legislation, for different reasons.
In this case the reasons seem twofold : firstly in France the parliament is still relatively free of aggressive lobbying from private interests, while the government is far more exposed to them. secondly, and more importantly, this particular piece of legislation is a transcription of European law. And national parliaments have no oversight of European law, which depends on the European Parliament and mostly on governments. And this particular piece of European law was actually passed at the time when the majority was opposite to what it was today.
Deputés like to rebel sometimes and try to pass amendments that they know go against the course set by the cabinet. But it's difficult for them because a strict voting discipline is enforced within the political parties, and dissent can have nasty consequences on a politician's career when the party leaders belong to the cabinet or are likely to in the future. And also their constituencies usually don't hold them accountable for their votes on individual issues. Instead they know they are expected to always vote along with the majority if they belong to it, or against if they are in the opposition.
It is a too-litteral translation of the French terms : "deputy" is for "député", who is a member of the lower house of parliement, the National Assembly. "government" here strictly means the executive branch of government, more precisely, the prime minister's cabinet (the president, even though he is part of the executive, is usually not considered part of what is covered by the word "gouvernement").
I'm an _atheist_ who occasionnaly uses the language expression "If God meant us to... (fly for example) then he would have (given us wings for example)". The irony is by that form of language I actually mean that we, or any species, are more comfortable when restricted to the circumstances that we have EVOLVED to deal with better.
Is more than 1.6 trillion USD, not 1.2. And that's using the purchasing power parity exchange rate, with which the euro is cheaper than its market rate. So at market rate that's even more when expressed in USD.
Slashdot Mirror
The US doesn't have "control of the internet", so it cannot be relinquishing what it doesn't have. The ICANN being US-based doesn't give much real control over IP packets travelling on some fiber halfway around the world from DC. Even if ICANN was a government agency it wouldn't. It just allows to vaguely arbitrate over domain names and IP number disputes that have relatively faint commercial implications. And even then the US feds would have to use indirect influence on ICANN.
I agree with the article that while India and China have been twin rising stars in the tales of ideological globalists like NYT's Thomas Friedman, there is a huge difference between the two : while China will be a superpower by the end of this century, at that time India will still be a third world country by far. With its caste structure, its irremediable lack of infrastructure and ressources to support its population, its relative submissiveness to western political pressure, the tendency of its educated elite to go live and work abroad the second they have a chance to, the best that can happen to India in the mid term is to nurture a developped sub-economy that will give it the global importance of, say, Italy, the UK or France.
Okay so that means the end of OS X within two years which brings a couple of benefits:
- Apple execs shouldn't be afraid to run out of cat names
- good for Linux since OS X was so far the only Unix embedded in a single, stable, nice GUI with drivers for all the gadgets in the market, it tended to divert Linux people who want their shell, their X11, their clean POSIX and their native Emacs/make.
They don't absolutely need these planes to be V/STOL since they are to be used on their future, full-size carrier which, unlike their existing mini-carriers will be able to operate any kind of naval jet like F/A-18 or Rafale M provided they are fitted with a catapult.
The design of their future carriers is already supposed to be largely shared with that of France's second carrier which is supposed anyway to operate Rafales. So the decision to dump V/STOL would simplify things actually.
Well they already have contracted out the design of the aforementionned carrier ships to a French company.
Plan B is naval version of the Typhoon, which is yet to be developped.
Plan C is the Rafale, whose naval version is already operationnal.
I don't think that the UK has ordered any JSF yet. What they did is pay part of the development costs, in return for the promise of a share of the industrial pie when the plane enters production. Their high level of financial participation was also supposed to buy them some input in the specification and some sharing of classified technologies, but the Americans largely didn't carry out that part of the deal, which has provoked transatlantic tension lately.
The JSF is supposed to equip the RNs future carrier ships around around 2015. However as a response to the US Congress looking at cancelation of plans for a Rolls-Royce engine equiped version of the JSF, the British have hinted that they could very well start developing a naval version of the Eurofighter Typhoon, or even consider the already operationnal naval version of the French Rafale.
I'm afraid this person who I assume is a fellow French national, is a little bit misinformed himself about the workings of his own country : the Cour de Cassation IS the highest court for civil law. There are three institutions that cover what the Supreme Court does in the US : The Cour de Cassation and the Council of State are the highest appeals for respectively civil law and administrative law. The former rules whether justice was administered properly in cases involving crimes and felonies of people or businesses. The later judges whether the State and the local governments act within their legit powers and is the highest appeal for people who, to put in in American terms, "sue the government". And finally the Constitutionnal Council censors bills when they are incompatible with the constitution or international law, _before thay are made into law_. It is not possible to appeal to it once the law is signed, unlike with the Supreme Court.
Yes : to break the ruling of a lower court. That's what the cour de cassation, which is the highest court for civil law, does. Note that the CdC does not rule on the content of cases, but only on the way justice was administered.
I don't think it's useless in the real world : first, complete formal validation of software is used in the real world for critical systems, even though it's considered anti-economical to use it for general programming. However it's more a matter of attitudes than of fundamental methodologies. And attitudes are shifting. I work at a semi-conductor company in the domain of formal hardware validation, and researchers here have noticed a significant academic brain drain from h/w validation towards formal software validation since Microsoft put its chips in it.
The Microsoft example is significant when I speak of "attitudes". The point is not to have every programmer mathematicaly prove properties on each and every function they write in an application (although core libs and OSs could be subject to that). The point is to have people with a _perspective_ of understanding what they write rather than an attitude of cutting-and-pasting "patterns" or algorithms from textbooks. What MS has stated is that they were doing this attitude shift.
If I had to interview people for a developper position, here's what I would do : I would ask them to write a "sort" function, preferably in whatever language they would be assigned to work in. And then I would review their code with them, not really paying any attention to whether the thing actually works or compiles or not. What I would ask them is "tell me why your code works". From there:
- VERY BAD ANSWER : "It works because look, I wrote a quicksort"
- QUITE BAD ANSWER ": "It works because it compiles and look, I give it this particular list and lo! I get a sorted list back."
- SOMEWHAT REASSURING ANSWER : "It works because I wrote this test unit and I analyzed the range of possible inputs and generated this stimulation set that covers all the border situations".
- WHEN ARE YOU AVAILABLE TO START ? : "It works because what I want is for the resulting list to contain all the entries in the input list, and only these, and that I want each element of the resulting list to be less or equal to its successor. When I call function qsort, as you can see here this particular line in my code means that the result will have its pivot in the right position, so by applying this invariant recursively, etc..."
I think that many people who are programming today, and even leaving code long term deep into production systems, should be kicked out of the field, when they don't really know what they're doing and have had a two years training of "programming-is-a-bit-like-spelling-out-a-cooking- recipe" style Java drill.
Programming requires an intellectual discipline, and an understanding of abstract mathematical concepts that is just not anybody's. Then we'll see systems working much better.
Well as I said, you have the wikipedia but I agree that doesn't sound very serious.
Let's try something else : The Lambda Calculus : its syntax and semantics by H.P. Barendregt, pages 131 and 132.
When I was a kid in the 70s, my parents were very enthusiastic in impregnating me with the technology that would later make me successful in life. They have really inspired my life and made my calling very clear early on in my life.
However right now I'm out of a job. Anyone needs someone who's REALLY GOOD at punching COBOL programs on cards ?
I don't know how intentionnal it is, I assume it is intentionnal : in computer science a Y combinator is, basically, a function that takes a function as an argument and returns a fixpoint to the function. so for any function f, f(Y(f))=Y(f). Oh, what's the hell : just check out the wikipedia.
My first computer was an EXL-100, from Exelvision of France, in 6th grade I believe which was, like, 1985. The company was started I believe by one or more French TI engineers who built their stuff from a TI chipset (rather than a single uproc). Its remarkable features were a "pro" form factor with a wireless (IR) keyboard separate from the CPU and some sound DAC used as a "voice synthetizer" giving pretty amazing results for the era.
One of a flurry of simple home computers from European companies in the early 80s, it didn't benefit from the government support (in the form of education sales) of its compatriots TO-7 and MO-5 from state-owned company Thomson. It didn't have the "cool games" and one year later I switched to a CPC-464 from Amstrad of the UK.
France does not have a pure civil law system, just as so-called "common law" countries don't have a pure precedent-based system. After all one of the sources of US law is a written source : the constitution.
In theory, the only law applicable in France is written by the legislative branch. The judiciary is supposed to only work as a machine, applying the following pseudo-code:
foreach (case) {
check out facts
look up relevant article of law
pronounce judgement
}
Precedent does exist, but is viewed more as a labor saver, a little bit like speeding up an algorithm by caching intermediate results that can be used more than once.
Most of the written law in France is arguably well designed, some of it, most famously Napoleon's civil code, having been barely touched in more than 200 years.
The benefit of a written law system then is that in many cases, if you want to know whether you're right to do something, you just look it up in the book (or on www.legifrance.gouv.fr), instead of giving $1000 to a lawyer for advice and crossing your fingers.
However some other codes are not working so well, two examples would be the intellectual property code, which is faced with a technology shift of millenary proportions, and the labor code, which is creaking under thousands of categorical-interest driven articles that are largely contradictory and contrarian to economic common sense.
In these domains, the precedents take center stage.
When I was a teenager in the '80s in France my parents rented a vacation home for two or three weeks, and as usual the owners had left a bunch of old books and magazines to garnish the bookshelves. Among them, tourism and travel guides to the USSR, which led my parents to frown upon having rented the place from people with communist sympathies. Browsing this guides was fascinating though, and I vividly remember how they "diplomatically" were introducing the potential traveller to the photography restrictions. Reading this, the notion of the authorities of a country restricting what could be photographed in it, particarly outdoors, sounded bizarre and impracticable...
So I don't get the logic behind the politically-correct Intel bashing. On the one hand, one hears that Intel is bad because they have carried on with binary compatibility since the microprocessor pre-history (the 8086, even the 8080 to a certain extent), creating an architecture that is indeed in some places a bit quaint.
Then on the other hand we are supposed to believe that AMD is genius for having led the way in moving to 64 bits through extension of that much-reviled x86 architecture rather than by starting from scratch with an innovative and modern architecture such as EPIC.
So the bottom line is the Intel is always wrong, right ?
Oh anyway, go and count how many www.top500.org supercomputers in the world are Intel powered, and how many are AMD powered. People who build supercomputing clusters must be clueless joe's who buy into Intel's evil marketing messages, right ? Because I see more "Itanics" than Opterons there, and way more Xeons than either of these.
"Military Police" is a very bad translation for Gendarmerie. The Gendarmerie is pretty much just a regular police force. Police in France is almost exclusively a national-level organization, unlike for example in the US where it is locally organized (city, county, state). Police is divided in two categories : the National Police, made of civil servants under the interior ministry, and the Gendarmerie, which is a branch of the military. Both do essentially the same job, wear similar uniforms, and are assigned to different districts, with the Police dealing with urban areas and the Gendarmerie with rural areas.
Differences are that the Gendarmes almost never work in plain clothes. They occasionaly wear olive green fatigues for field work away from civilian contact. They can be sent overseas as easily as the other military people, which is very convenient for France in providing trained civilian-style law enforcement capabilities in international peace keeping missions.
Other countries have or have had similar forces : the Gendarmerie in Belgium, the Guardia Civil in Spain, both of which I believe have been "civilianized". The Carabinieri in Italy.
It's hasty translation. "Gouvernement" in France means "Prime Minister's cabinet", which, along with the presidency, form what Americans would call the executive branch of government.
The Prime Minister and his cabinet are always according to the majority elected to the lower house of parliament, which is not the case of the president who is elected independently.
However the MPs (députés), even though they theoretically can draw up bills or reject the gouvernement's proposed ones, usually acts as a rubber stamp to the cabinet designed legislation, for different reasons.
In this case the reasons seem twofold : firstly in France the parliament is still relatively free of aggressive lobbying from private interests, while the government is far more exposed to them. secondly, and more importantly, this particular piece of legislation is a transcription of European law. And national parliaments have no oversight of European law, which depends on the European Parliament and mostly on governments. And this particular piece of European law was actually passed at the time when the majority was opposite to what it was today.
Deputés like to rebel sometimes and try to pass amendments that they know go against the course set by the cabinet. But it's difficult for them because a strict voting discipline is enforced within the political parties, and dissent can have nasty consequences on a politician's career when the party leaders belong to the cabinet or are likely to in the future. And also their constituencies usually don't hold them accountable for their votes on individual issues. Instead they know they are expected to always vote along with the majority if they belong to it, or against if they are in the opposition.
It is a too-litteral translation of the French terms : "deputy" is for "député", who is a member of the lower house of parliement, the National Assembly. "government" here strictly means the executive branch of government, more precisely, the prime minister's cabinet (the president, even though he is part of the executive, is usually not considered part of what is covered by the word "gouvernement").
I'm an _atheist_ who occasionnaly uses the language expression "If God meant us to ... (fly for example) then he would have (given us wings for example)". The irony is by that form of language I actually mean that we, or any species, are more comfortable when restricted to the circumstances that we have EVOLVED to deal with better.
Then all biologists should be charged with violating the DMCA, shouldn't they ?
Is more than 1.6 trillion USD, not 1.2. And that's using the purchasing power parity exchange rate, with which the euro is cheaper than its market rate. So at market rate that's even more when expressed in USD.