The main difference in modern, non political-sciences usage is that a republic has a directly elected head of state, while a parliamentary democracy have a head of state elected by the parliament. Many constitutional monarchies fall in the latter, though in some cases the official head of state will be the monarch, though with practically no real power. Both republics and parliamentary democracies are subsets of what most people call democracies in general usage.
Some will insist, though, that democracy is a term that should only be used about direct rule. For those people, I'd suggest they go read "After the revolution? Authority in a good society" by professor Robert A. Dahl, Yale University Press. It's an easy read, and will make you think about the problems of authority and democracy, and why modern democracies are the way they are.
First, as others have pointed out, it is NOT true that the US have had the longest continuous democratic government in the world - other countries have had you beat by centuries.
Even discounting that, the US also has not been leading in terms of who can vote: During the early days of the US, voting restrictions "even" on men were strict - many states requiring property ownership etc.. It wasn't until the 1840's that restrictions on voting for Catholics and non-Christians were lifted. Black males didn't get the right to vote until the 15th amendment in 1870, but the right was severely abridged by states putting in place requirements that were hard or even impossible to meet ("grandfather clauses" giving voting rights only to peoples whose grandparents had the right to vote etc.) - most of these restrictions lasted at least until 1915. Women didn't get the right to vote at federal level until 1920, solidly beat both 18-19 countries at least. Indians didn't get the right to vote until 1924.
The final restrictions designed to prevent blacks from voting didn't disappear until poll taxes were made illegal in the 60's
So how do you define "democratic government"?
To compare, New Zealand was the first country in the world to introduce universal voting rights - from 1893 women where given the right to vote, 23 years after the indigenous population got their right to vote.
If you, however, accept all the restrictions above, as "democratic", then the list of countries with older continuous democratic governments than the US is quite long.
The code is already beeing written for you. It virtually "merges" two directory hierarchies. If you want to see the separate trees, you just look at the original locations.
Re:acid was eating away at the reactor cover shiel
on
China Goes Nuclear
·
· Score: 1
I haven't bothered reading most of your silly rant, but I'll point out two things:
- The "Chinese Communist experiment" is nothing of the sort. China tried to force an agrarian society into socialism, screwed up (as they should have expected from day one given how it failed in Russia, and given that this is exactly what Marx' claimed would happen to under developed countries that attempted socialism), and started moving towards a market economy. Today it has one of the largest private company based economies in the world, and is quickly privatising more and more of the state companies.
- Big government failing is a central theory of communism. Go read "State and the revolution" by Lenin, which very clearly summarizes the marxist theory that the state is an apparatus for class oppression and that a transition to comunism would make it necessary for the state to "wither away" and be at most maintained only in the form of an organisation with purely administrative support functions.
Looking at Marxism, there is even nothing contradictory between communism and a free market (if you're looking to Leninism or Maoism, then yes, they were both very much about planned economies, but then they've failed too). The difference being between a capitalist society with private ownership of the means of production and a communist society with public ownership of the means of production.
Competition about a free market is about much more than profit - it is also about survival and keeping control of your destiny, which is why so many people run their own businesses even when they could have made wastly more money by working for a large corporation in a less stressfull position. The money factor would to some extent be absent in a communist free market system, but the remaining factors that cause people to want to run their own companies and compete would still be present.
Yeah, well, they didn't test run the Titanic into an iceberg before embarking on the maiden voyage, though, so I have a little more faith in this project. If they are sure enough about the concept to be prepared to to multiple tests of shutting down the cooling it seems promising. Particularly since the principle has been known since the 50's.
Read the article. One of the great things about the Chinese reactor is that the fuel balls are relatively effectively contained and can be stored "safely" with very little extra shielding. It's still going to be a problem long term, but if their claims have merit it sounds like a major step towards making nuclear waste into a manageable problem.
It hasn't taken the Chinese this long to announce this - it's taken Wired this long to write an article about it. China has had nuclear power plants for years. The "news" here (which are old) is that China is planning to massively increase the number of plants, and secondly that they have a working prototype of what they claim is a meltdown safe reactor. The Wired article is good, but most of the stuff on China's build out of reactors was background material for covering their pebble bed reactor.
Re:Nuclear power plants as strategic targets
on
China Goes Nuclear
·
· Score: 1
Maybe, but do you have any indication that it will kill more people than blowing up an "easier" target like a skyscraper? Traditional reactors are built to be solid enough to mostly contain a complete meltdown. That means A LOT of protective materials. They are also high security areas - the chance of getting explosives close enough (or getting to somewhere where you could trigger a meltdown from) would be a lot harder. The new reactor types China are planned on the other hand would be unlikely to cause much damage - they might increase radiation levels, but given that they're intended to be much smaller (both in terms of total size and amount of fuel) the potential for massive radiation leaks should be smaller as well unless you'd manage to blow up a series of linked reactors.
Re:China's Fusion... Or lack thereof.
on
China Goes Nuclear
·
· Score: 1
The Great Wall story you link to has little to do with China telling lies. Their first astronaut didn't see the wall, but at least one US astronaut claims to have seen it.
It's also doubtful if that story originated in China in the first place, and China haven't had any possibility of verifying it.
There's also a fundamental difference between the nuclear project and the moon base gaffe you link to: In the case of the moon base claims, a single official made a claim that was retracted publicly soon after they were published. In the case of the nuclear projects, it is official Chinese policy, backed by multiple reports that indicate it is an absolute necessity, and is already subject to negotiations with some of the companies that have already delivered nuclear reactors to China.
If you RTFA you'll see that the South African and German programs are both covered. The main difference is that the Chinese model is meltdown proof (or so they claim anyway) because they keep the reactor core small enough that it will cool down by itself.
It is now the result of a script that is the result of a persons creativity. What difference does it make? What matters is the result, and the result will still depend on the composers ability to manipulate the music, whether that is done by playing directly on an instrument or modifying a program to change the way the program plays the instruments.
Besides composition has always been highly rule based, and even classics like Mozart toyed with generative music. There's a lot of music out there that is clearly composed in ways that closely resemble many of the methods used by generative music systems today. Beethovens "Ecossaisen" for instance closely follow the structure of a Cantor fractal. Composing music IS about combining patterns much more than about combining individual notes.
This page is a good overview of the fractal nature of composition.
I'd agree with you that using a computer for generating the music for a live performance would be a cop out if the computer was programmed with the full score from the start, instead of being used as an instrument that allows automation of repetitive elements.
Personally I see very little value in live performances of the actual music if the music is intended to just follow a score - what makes a live performance interesting to me is what goes into the presentation and/or any improvisation, not that someone demonstrates that they are capable of recreating the same tunes over and over in realtime.
Their explanation is crap. I'm sure I'll make tons of mistakes and get flamed to death by people who actually paid attention when this was covered at school, but here's an attempt at a more complete explanation...
A "P" problem is a decision problem (a yes/no question essentially) that can be solved in polynomial time: Worst case for searching for a book in a finite collection is a multiple of the worst case time it takes to search one book in the house and the number of books - a straight linear search of all books in the house.
An "NP" problem is a decision problem where a solution can be verified in polynomial time. If P=NP then any problem where a solution can be verified in polynomial time can also be solved in polynomial time. Solving the problem here refers to finding the answer (is the book I seek in the house?) while verifying the solution refers to controlling that the answer is correct.
An P problem is also an NP problem - if you can solve a problem in polynomial time you can by definition also verify an answer in polynomial time: solve the problem, and compare the result with the proposed answer.
However the problem is that while it is thought that not all NP problems are also in P, it is not proven, and may be unprovable. But we don't know which, and don't have a simple way of verifying whether an NP problem is in P or not.
This is what sucks for encryption. Encryption currently rely on it being hard to solve a specific problem but easy to verify the solution (the key) by trying to decrypt an encrypted message, and if it fails (or presents us with gibberish) we know the key presented to us was not the right solution. This can be done in polynomial time
However, it only protects us because nobody (at least as far as we know) can solve the problems posed by the encryption methods in less than exponential time. Exponential time to solve the problem means that the effort could for instance (but not necessarily) double for each extra bit added to the key. The general rule is that the effort is 2^p(n) where p() is a polynomial function of n, and n is the size of the information space of the problem. The problem asked by encryption is essentially "will this key slotted into this encryption function give the original message?"
Let's go back to the book search, and twist it so that it fits the situation with encryption now. In the book search, the question is "does the book I seek exist in the house?". In a normal world, this problems worst case solution increases in effort by a constant factor for each extra book added, but to make it comparable to encryption we need to complicate the problem.
Imagine a crazy world where the best way we have to look for a solution to the book problem is to compare all the books with eachother and a paper strip with the title of the book we want, and only when we happen to compare the paper slip with the right book have we found the solution. In that case, each time we add a book, the problem increases (the number of direct relationships between n entities n(n-1), where n in this case would include the paper strip with the solution) but where we can still verify the final solution by comparing the book we finally found with the paper strip.
The kind of twisted scenario above is what most encryption algorithms are today: A problem that gives protection only because we don't know how to quickly solve it, but where there could be a simple way of solving them right around the corner. If P=NP all encryption algorithms that depend on this relationship can be broken in polynomial time.
However even proving that P and NP are not equal only means that "safe" encryption using this relationship is possible, not that current algorithms are safe - to prove that an algorithm is safe you'd need to separately prove that the problem can only be solved in exponential time. But a proof that P!=NP would give us an indication that it might at least be possible for such a proof to exist, as proving that the problem posed by an encryption algorithm can only be solved in exponential time would at the same time prove that P!=NP.
No it's not a funny claim to make. One of the defining features of a supercomputer vs. a cluster is tightly intergrated low latency high bandwidth interconnects, whereas clusters typically rely on software support (networking etc.) to shuffle data over relatively low speed, high latency communications channels.
And in doing so you are essentially building a super computer. However you'd have to keep in mind that it isn't all about total bandwidth - latency also needs to be extremely low. That said, HP is working on an open source Single System Image clustering support for Linux on "normal" hardware
Cray supports Linux. In fact, the supercomputer platform Cray's CTO was pushing in the interview is running Linux. What he's saying is really that supercomputers can handle classes of problems that clusters have problems with.
If your goal is to run simulations where each piece of the simulation depend on large subset of the other pieces, then you will need ridiculous interconnect speeds, and you're likely to end up with something you could have bought from Cray or SGI or some of the other remaining supercomputer manufacturers for a fraction of the price.
Luckily for you and the rest of us many problems can be split into relatively independent pieces, in which case a Beowulf cluster or similar is more than adequate.
If you seriously believe that clusters can compete with supercomputers for every type of problem, you need to think again.
It's too easy to assume that the reason they have problems with the concepts is that they don't have the words - the other way around is frequently the case: you won't have the words if you don't have the concepts, or if cultural differences means you have never had a need to express something.
One of my favourite examples, as a Norwegian stranded in the UK, a country where people simply does not get the concept of candy with ammonium chloride, is how to talk about it.
In the UK, the word "candy" has mostly gone out of use, and usually refers to brown sugar or alt least "old fashioned" sweets based on brown sugar. Instead you'd refer to the different types of confectionary directly, with most of the sugar based confectionary grouped under "sweets".
Now, ammonium chloride based candy is most definitively not sweets. Though it is always fun to trick Brits into chewing Turkish Pepper or some other Scandinavian ammoium chloride based candy...:)
The word "confectionary" similarly doesn't really cut it - it's recognised as a grouping, and if you asked people if thy wanted any confectionary they'd wonder what kind you were talking about.
Scandinavian languages on the other hands have words for this, since it's an integral part of our culture. In Norwegian you'd talk about "godt" or "smaagodt", referring to small sweets, bits of licorice, small chocolate pieces or candy full of ammonium chloride, as well as assorted sour stuff.
But what would a usable equivalent be in the UK? I usually end up resorting to candy, but Brits then tend to assume that since I'm foreign I'm probably resorting to US English, and talking about sweets...
RTFA. The internal reports does NOT support the claims that it was life support, only that it affected their internet connectivity. From the securityfocus article:
"Given the fact that no financial records or systems were compromised, no safety or loss of life was threatened, and no critical system corrupted, we need to balance legitimate security needs with the legitimate needs of our scientists at the Pole," the memo reads.
Using the term "infiltrate" for placing persons in an organisation where membership is open to the public is perfectly normal when the persons in questions are purposefully joining with an agenda that is at odds with the goals of the organisation for the purpose of obtaining information that would not otherwise be readily available (what goes on at meetings that is limited to the organisations members, for instance) or affect the way the organisation operate.
Brokerages are required by the SEC to ensure that you have at least a basic understanding of the risks associated with riskier types of investment like IPO's, options, margin trading etc.. If these reporters answered "wrong" (for instance told their broker their long term goal was steady growth and that they were looking for a profit within the next 5 years - something which would be a clear indication that an IPO wouldn't be the right investment) the broker would
have to reject them.
This is standard practise, and a reporter that didn't even know something as basic as that perhaps wasn't the right reporter to follow a story about an IPO.
I think the point was that what matters for a company that wants to sell their products worldwide is political knowledge more than geography or history - some information is simply not palatable to a lot of governments or communities regardless of whether it's accurate or not.
Slashdot Mirror
Some will insist, though, that democracy is a term that should only be used about direct rule. For those people, I'd suggest they go read "After the revolution? Authority in a good society" by professor Robert A. Dahl, Yale University Press. It's an easy read, and will make you think about the problems of authority and democracy, and why modern democracies are the way they are.
Even discounting that, the US also has not been leading in terms of who can vote: During the early days of the US, voting restrictions "even" on men were strict - many states requiring property ownership etc.. It wasn't until the 1840's that restrictions on voting for Catholics and non-Christians were lifted. Black males didn't get the right to vote until the 15th amendment in 1870, but the right was severely abridged by states putting in place requirements that were hard or even impossible to meet ("grandfather clauses" giving voting rights only to peoples whose grandparents had the right to vote etc.) - most of these restrictions lasted at least until 1915. Women didn't get the right to vote at federal level until 1920, solidly beat both 18-19 countries at least. Indians didn't get the right to vote until 1924.
The final restrictions designed to prevent blacks from voting didn't disappear until poll taxes were made illegal in the 60's
So how do you define "democratic government"?
To compare, New Zealand was the first country in the world to introduce universal voting rights - from 1893 women where given the right to vote, 23 years after the indigenous population got their right to vote.
If you, however, accept all the restrictions above, as "democratic", then the list of countries with older continuous democratic governments than the US is quite long.
Why?
The code is already beeing written for you. It virtually "merges" two directory hierarchies. If you want to see the separate trees, you just look at the original locations.
With a father like that, I'm not at all surprised he has issues...
Yay! That mean I can finally get the personal nuclear reactor for my new house as all the SF stories promised.
The "myth" was first presented in a book in 1938, before the PRC even existed, and at least one US astronaut has claimed to have seen it with the naked eye from LEO.
- The "Chinese Communist experiment" is nothing of the sort. China tried to force an agrarian society into socialism, screwed up (as they should have expected from day one given how it failed in Russia, and given that this is exactly what Marx' claimed would happen to under developed countries that attempted socialism), and started moving towards a market economy. Today it has one of the largest private company based economies in the world, and is quickly privatising more and more of the state companies.
- Big government failing is a central theory of communism. Go read "State and the revolution" by Lenin, which very clearly summarizes the marxist theory that the state is an apparatus for class oppression and that a transition to comunism would make it necessary for the state to "wither away" and be at most maintained only in the form of an organisation with purely administrative support functions.
Looking at Marxism, there is even nothing contradictory between communism and a free market (if you're looking to Leninism or Maoism, then yes, they were both very much about planned economies, but then they've failed too). The difference being between a capitalist society with private ownership of the means of production and a communist society with public ownership of the means of production.
Competition about a free market is about much more than profit - it is also about survival and keeping control of your destiny, which is why so many people run their own businesses even when they could have made wastly more money by working for a large corporation in a less stressfull position. The money factor would to some extent be absent in a communist free market system, but the remaining factors that cause people to want to run their own companies and compete would still be present.
Yeah, well, they didn't test run the Titanic into an iceberg before embarking on the maiden voyage, though, so I have a little more faith in this project. If they are sure enough about the concept to be prepared to to multiple tests of shutting down the cooling it seems promising. Particularly since the principle has been known since the 50's.
Read the article. One of the great things about the Chinese reactor is that the fuel balls are relatively effectively contained and can be stored "safely" with very little extra shielding. It's still going to be a problem long term, but if their claims have merit it sounds like a major step towards making nuclear waste into a manageable problem.
It hasn't taken the Chinese this long to announce this - it's taken Wired this long to write an article about it. China has had nuclear power plants for years. The "news" here (which are old) is that China is planning to massively increase the number of plants, and secondly that they have a working prototype of what they claim is a meltdown safe reactor. The Wired article is good, but most of the stuff on China's build out of reactors was background material for covering their pebble bed reactor.
Maybe, but do you have any indication that it will kill more people than blowing up an "easier" target like a skyscraper? Traditional reactors are built to be solid enough to mostly contain a complete meltdown. That means A LOT of protective materials. They are also high security areas - the chance of getting explosives close enough (or getting to somewhere where you could trigger a meltdown from) would be a lot harder. The new reactor types China are planned on the other hand would be unlikely to cause much damage - they might increase radiation levels, but given that they're intended to be much smaller (both in terms of total size and amount of fuel) the potential for massive radiation leaks should be smaller as well unless you'd manage to blow up a series of linked reactors.
It's also doubtful if that story originated in China in the first place, and China haven't had any possibility of verifying it.
Wikipedia claims the story originated from Richard Halliburton in 1938, before the PRC was even established.
There's also a fundamental difference between the nuclear project and the moon base gaffe you link to: In the case of the moon base claims, a single official made a claim that was retracted publicly soon after they were published. In the case of the nuclear projects, it is official Chinese policy, backed by multiple reports that indicate it is an absolute necessity, and is already subject to negotiations with some of the companies that have already delivered nuclear reactors to China.
If you RTFA you'll see that the South African and German programs are both covered. The main difference is that the Chinese model is meltdown proof (or so they claim anyway) because they keep the reactor core small enough that it will cool down by itself.
Besides composition has always been highly rule based, and even classics like Mozart toyed with generative music. There's a lot of music out there that is clearly composed in ways that closely resemble many of the methods used by generative music systems today. Beethovens "Ecossaisen" for instance closely follow the structure of a Cantor fractal. Composing music IS about combining patterns much more than about combining individual notes.
This page is a good overview of the fractal nature of composition.
I'd agree with you that using a computer for generating the music for a live performance would be a cop out if the computer was programmed with the full score from the start, instead of being used as an instrument that allows automation of repetitive elements.
Personally I see very little value in live performances of the actual music if the music is intended to just follow a score - what makes a live performance interesting to me is what goes into the presentation and/or any improvisation, not that someone demonstrates that they are capable of recreating the same tunes over and over in realtime.
A "P" problem is a decision problem (a yes/no question essentially) that can be solved in polynomial time: Worst case for searching for a book in a finite collection is a multiple of the worst case time it takes to search one book in the house and the number of books - a straight linear search of all books in the house.
An "NP" problem is a decision problem where a solution can be verified in polynomial time. If P=NP then any problem where a solution can be verified in polynomial time can also be solved in polynomial time. Solving the problem here refers to finding the answer (is the book I seek in the house?) while verifying the solution refers to controlling that the answer is correct.
An P problem is also an NP problem - if you can solve a problem in polynomial time you can by definition also verify an answer in polynomial time: solve the problem, and compare the result with the proposed answer.
However the problem is that while it is thought that not all NP problems are also in P, it is not proven, and may be unprovable. But we don't know which, and don't have a simple way of verifying whether an NP problem is in P or not.
This is what sucks for encryption. Encryption currently rely on it being hard to solve a specific problem but easy to verify the solution (the key) by trying to decrypt an encrypted message, and if it fails (or presents us with gibberish) we know the key presented to us was not the right solution. This can be done in polynomial time
However, it only protects us because nobody (at least as far as we know) can solve the problems posed by the encryption methods in less than exponential time. Exponential time to solve the problem means that the effort could for instance (but not necessarily) double for each extra bit added to the key. The general rule is that the effort is 2^p(n) where p() is a polynomial function of n, and n is the size of the information space of the problem. The problem asked by encryption is essentially "will this key slotted into this encryption function give the original message?"
Let's go back to the book search, and twist it so that it fits the situation with encryption now. In the book search, the question is "does the book I seek exist in the house?". In a normal world, this problems worst case solution increases in effort by a constant factor for each extra book added, but to make it comparable to encryption we need to complicate the problem.
Imagine a crazy world where the best way we have to look for a solution to the book problem is to compare all the books with eachother and a paper strip with the title of the book we want, and only when we happen to compare the paper slip with the right book have we found the solution. In that case, each time we add a book, the problem increases (the number of direct relationships between n entities n(n-1), where n in this case would include the paper strip with the solution) but where we can still verify the final solution by comparing the book we finally found with the paper strip.
The kind of twisted scenario above is what most encryption algorithms are today: A problem that gives protection only because we don't know how to quickly solve it, but where there could be a simple way of solving them right around the corner. If P=NP all encryption algorithms that depend on this relationship can be broken in polynomial time.
However even proving that P and NP are not equal only means that "safe" encryption using this relationship is possible, not that current algorithms are safe - to prove that an algorithm is safe you'd need to separately prove that the problem can only be solved in exponential time. But a proof that P!=NP would give us an indication that it might at least be possible for such a proof to exist, as proving that the problem posed by an encryption algorithm can only be solved in exponential time would at the same time prove that P!=NP.
The formats doesn't mean anything. The only thing that matters are: low price, "good enough" quality, and what content is pushed out on each format.
No it's not a funny claim to make. One of the defining features of a supercomputer vs. a cluster is tightly intergrated low latency high bandwidth interconnects, whereas clusters typically rely on software support (networking etc.) to shuffle data over relatively low speed, high latency communications channels.
And in doing so you are essentially building a super computer. However you'd have to keep in mind that it isn't all about total bandwidth - latency also needs to be extremely low. That said, HP is working on an open source Single System Image clustering support for Linux on "normal" hardware
If your goal is to run simulations where each piece of the simulation depend on large subset of the other pieces, then you will need ridiculous interconnect speeds, and you're likely to end up with something you could have bought from Cray or SGI or some of the other remaining supercomputer manufacturers for a fraction of the price.
Luckily for you and the rest of us many problems can be split into relatively independent pieces, in which case a Beowulf cluster or similar is more than adequate.
If you seriously believe that clusters can compete with supercomputers for every type of problem, you need to think again.
One of my favourite examples, as a Norwegian stranded in the UK, a country where people simply does not get the concept of candy with ammonium chloride, is how to talk about it.
In the UK, the word "candy" has mostly gone out of use, and usually refers to brown sugar or alt least "old fashioned" sweets based on brown sugar. Instead you'd refer to the different types of confectionary directly, with most of the sugar based confectionary grouped under "sweets".
Now, ammonium chloride based candy is most definitively not sweets. Though it is always fun to trick Brits into chewing Turkish Pepper or some other Scandinavian ammoium chloride based candy... :)
The word "confectionary" similarly doesn't really cut it - it's recognised as a grouping, and if you asked people if thy wanted any confectionary they'd wonder what kind you were talking about.
Scandinavian languages on the other hands have words for this, since it's an integral part of our culture. In Norwegian you'd talk about "godt" or "smaagodt", referring to small sweets, bits of licorice, small chocolate pieces or candy full of ammonium chloride, as well as assorted sour stuff.
But what would a usable equivalent be in the UK? I usually end up resorting to candy, but Brits then tend to assume that since I'm foreign I'm probably resorting to US English, and talking about sweets...
"Given the fact that no financial records or systems were compromised, no safety or loss of life was threatened, and no critical system corrupted, we need to balance legitimate security needs with the legitimate needs of our scientists at the Pole," the memo reads.
Using the term "infiltrate" for placing persons in an organisation where membership is open to the public is perfectly normal when the persons in questions are purposefully joining with an agenda that is at odds with the goals of the organisation for the purpose of obtaining information that would not otherwise be readily available (what goes on at meetings that is limited to the organisations members, for instance) or affect the way the organisation operate.
This is standard practise, and a reporter that didn't even know something as basic as that perhaps wasn't the right reporter to follow a story about an IPO.
I think the point was that what matters for a company that wants to sell their products worldwide is political knowledge more than geography or history - some information is simply not palatable to a lot of governments or communities regardless of whether it's accurate or not.