And, in this discussion, very important to note is that the US had a very big hand in setting up the German system this way. The US election problems are not recent, the politicians really knew the faults, as far back as 1945.
Why are Ohio, Pennsylvania, New Mexico, etc., or whatever state that's in the swing more important than the rest of the US? With a proportional system, every vote counts. With a district system, all regions count. Why is it good that a republican minority in New York doesn't get any representation, the same goes for a democratic minority in Idaho? Winner takes all seems decisive, but it's not all that fair, especially not on the national level. Maybe take a look at a true modern democracy, the democracy of Germany (invented by the US), and see how both regional and national interests are balanced in the Bundestag. You might learn from yourselves someday...
Given the normal distribution, if they're 95% sure it's within 3% of the stated number (thus plus or minus two (1.96) standard deviations), they are 99% sure that it's within 4.2% of the stated number. To get a difference larger than 5.6% this would be 1 in a 1000 (0.999). 8% would be once in a million. In jargon: normal tails die of quickly.
More accurately, statistics is the science of the inexact, and actually a branch of mathematics. Any experiments done in science has statistical results, inexact results, where the inexactness is quantified. Statistics is about two main things: bias and variance. Bias occurs when either a theory is wrong, or the respondents in the theory, be it electrons or voters, refuse to give their readings. If it's done in a systematic way, it's called bias. Variance is a general uncertainty around this bias. Bias can be quantified, for instance that conservatives are less likely to respond to the poll. This can, and will, be catered for. Variance caters for the scaling up: if you queried 500 people, and try to scale up to the entire population, you can quantify this by taking the number of respondents into account. If both the bias and the variance don't match up with an election result, i.e., the calculation show that the actual outcome is improbable both with the bias (i.e., the expected outcome, catering for systematic error) and the variance (the uncertainty around this estimate), one of two things can be the matter: your calculation of bias is wrong, or the system is rigged. Blaming it purely on the statistics is very easy, but not necessarily correct.
By publishing and researching the bias calculations you might be able to figure out which of the two is the case. Such a thing has not been done, or, at least, not been published.
In the end this leads to a whole lot of unanswered questions, meaning that the results are, and will be, in doubt. Taking care of this doubt should be a number one issue in running any new election, trying to reason it away is not helpful. In most countries it is accepted that the vote will be off by quite a bit, only in the US this is the difference between having one administration or a different one. Please observe Brazil at this moment. They are going into a second round because the incumbent didn't get a decisive vote. In France, the difference needs to be 5 percent. This takes care of a lot of error due to variance (due to unintentionally making the wrong vote), and the counting process. My opinion on this is that the oldest modern democracy in the world is in a crisis and needs to re-examine its process.
"I think therefore I am" is a statement of observation and reasoning, a very famous one. Even if the observation 'I think' was an assumption (which it isn't, how else can you observe something by thinking it. Think about this), this would not be faith, but belief: two totally different things. I am an atheist, I do not have faith. I however believe many things, including that the sun comes up each morning. This is not faith, this is a belief (and a bloody well founded one as well). Disentangling the mess that religion has made of the terms faith and belief is a step further towards understanding.
I'm sorry, but this argument was also in response to the growing, and in this thread expressed, opinion that people *outdoors* should also refrain from smoking. Outdoors is a public place, and that was what I was aiming at. People are defining public places to include parks, the outside of bars, and other places. This is were my comments were aimed at. I am perfectly happy with not smoking in confined areas as I do see the point in not forcing my habit onto others, I am however dead against extrapolating that to outdoor activities. I'm reading alarming stories these days of health-fundamentalists to do just that.
But, in any case, I do live in a city, and I am forced to inhale car fumes of people that enjoy their mobility for no good reason. What are your thoughts on this, oh man with the swinging fist (*). You might enjoy your meal without being forced to inhale cigarette smoke, I on the other hand cannot enjoy a walk without the great stimulant of car fumes. Many studies have shown that kids growing up in cities have more health problems than kids in the country. Why is your mobility more important than my health?
(*) I'm actually six foot three, 210 pounds. I would love you taking a swing at me;)
I start with the assumption that there's a 50% chance that the sun will go up tomorrow: particularly, I start with the (Laplacian) prior that I have one observation that the sun comes up, and one observation that the sun doesn't come up. Every day that passes, I add one to the observation that was true: in this case invariably that the sun indeed does come up. After 10 years of observations, I'm concluding that there's a 10*365/ (10*365 + 1) = 0.99973 probability that the sun indeed comes up the next morning. If I take into account that I believe that the historical records are not lying and that we had millenia of physical laws not changing, this probability is a severe underestimate.
In public, who the fuck are you to force me to inhale your car's pollutants? I suggest to put a stop to that and create an intricate system where somebody has to prove, and pay dearly, that there is no conceivable other way of transportation *every time* they take a car. What the fuck gives you the right to pollute thousands?
And with such an extensive database it's not that impossible to unwillingly write an entire paragraph 'ipsis verbis' to some previous paper.
Try to do the math and see where it breaks down. Vocabulary of students: 10,000 words (pretty low), number of possible 10 word sentences (short paragraphs): 10,000^10=1e40. Percentage of nonsense sentences due to grammaticality constraints: 0.99999 (wild guess, but add extra 9's if you so desire. Also consider level of knowledge of grammar by high-school students.). Leaves 1e35 possible paragraphs. Even if I'm 20 orders of magnitude off, the 22 million papers do not even make a dent in this space.
This is quite a wild guess, but we can also tackle it emperically. Take a random website with some text on it: take 10 consecutive words of that website and feed that to google (using quotes to make sure that you only get exact matches). Now count the results. You will either find your original back, or another website that has plagiarized your original (or vice-versa).
Just as a test, I took a message from the kde-man viewer:
"Be careful that you must take care about upper case and lower case characters". Pretty innocuous right? Somebody might have made this particular grammatical construct before. Wrong. It's unique. Check the results.
Now try to write your own sentences of 10 words and see if you can get a hit. For instance: "Make sure that you only use upper and lower case characters". No, no hits. "Does rewriting a paper you found in your own words also count?" No hits. Maybe take your entire post and feed it, 10 consecutive words at a time, to Google. I'd be suprised if you get a single hit, given that your post hasn't been cached yet. You are truly a unique person that produces unique writing, even 10 words at a time!
The public school might argue that the reason for keeping the copyright can only be to allow someone to copy your work later, i.e., cheating. Thus, students wanting to keep copyright on their work are assumed to help with cheating, and will therefore fail the class.
In all honesty, I think the schools would be better off to put all old papers on a search engine-accessible part of their own website, and do an automatic check using whatever search engine that does the job (read: Google) for plagiarism. That way, there's no commercial entity involved that swallows student's work, and if many schools do this, they get the same network effect as with turnitin.
As for drugs, non-violent drug use has victims too. The broken homes and families where someone becomes an addict and ruins their life. I'm quite happy to spend my tax dollars for prisons and cops.
You're absolutely right: drug use has victims. However, the prisons and cops have probably destroyed more families (dad in prison for smoking), than they have saved. At a huge cost to society. And then there's still alcohol for the law-abiding citizen: now that's a home-wrecker exceeding most other things. Let alone the Colombian farmer whose livelihood gets destroyed to officially save these poor Americans from themselves as if these can't be trusted to lead their lives the way they want to. All very pathetic, really.
I love this country and I'd gladly give up privacy to ensure that my children are safer.
And what evidence gives you this ridiculous notion that giving up privacy ensures that your children are safer? What basis has your belief that your government is only interested in your best interests and will not misuse the freedoms you give them?
So on the one hand you have the unproven assertion that increased spying will actually gain security (*), while on the other hand you have the historical certainty that powers given to the government will be abused to the full extent of the law (**). Your children will be proud of you.
(*) Note that there was sufficient intelligence to prevent 9-11. It was just not used.
(**) And already librarians, Quakers and knitting clubs are being harassed.
Given that this particular State's officials might actually be in power exactly due to the interferebility of the Diebold machines, why do you think they'll endanger their reelection by throwing out Diebold?
No, for this discussion, you're living in a representative democracy. The fact that it's a republic is irrelevant, as many constitutional monarchies testify. If you want to be pedantic, please be precise (*).
(*)yes, the second lemma of many dictionaries in the US do state that a republic is a form of representative democracy, yet the first lemma is invariably a 'country that has a president (elected or not) as head of state'. Indeed this does seem to sum up the breadth of democracy the US is experiencing now, but I don't think that's what you meant.
Please explain why it's better for the Dems to recount only in strong Dem counties. The counties already have many Dem votes (them being Dem counties and all), so it seems there are a lot less shenanigans to be performed in such counties. You'd better shenanige in Reb counties as you've got many more Reb votes to shen to the Dem side.
Or could it be that the Dem county Dem vote was so much lower than expected that they tried to fix that one first, given that any irregularities would have been tried there?
Luckily SCOTUS decided that it was way more important to have a president before the 15th of January than to figure out who should be Prez. Can't blame them, any figurehead will do: it's following the rituals of Democracy that make a country Democratic.
Computer monocultures are inherently unstable and when corrupted once, the mono-culture can be brought down exponentially fast. This holds for any mono-culture in computing: security software not excluded.
If I want any office-type of job, whether computer-related or simply sales, I'd better know how to operate windows/office, otherwise: no job. If I want to be able to work from home from time to time, I'd better have at least one windows/office machine at home, whether company paid or by myself. If I choose to use OpenOffice, I'd better be humble when something goes wrong in cooperating. Although it is accepted that office fucks up itself from time to time, it's not accepted of any other app. By the job. The big monopoly of Microsoft is in the workplace, not at the homes.
The choice for or against MS is simple for many: job or no job. Is that enough of a monopoly for you?
Probably Google send the newspapers the instructions on how to prevent being cached, and considered the case done. If they insist shooting themselves in the foot, let it be.
hmm, reading further in the thread, apparently my lack of knowledge of the syntax of robots.txt is only matched by my willingness to make a fool of myself. If indeed NOCACHE is an option, and respected by Google, then these Belgians are stupid to go to court for something that has an easy technological fix and the court is stupid to allow this.
Can you use robots.txt for this? The situation is this: I have an article in/public/article.html, and later move it into my archive at/archive/article.html, where only paid subscribers can see it. My robots.txt dutifully notices that/archive is off-limits. How to tell Google to delete their cached version of article.html?
Apparently only by making/public offlimits as well. This is not the intention.
Slashdot Mirror
And, in this discussion, very important to note is that the US had a very big hand in setting up the German system this way. The US election problems are not recent, the politicians really knew the faults, as far back as 1945.
Why are Ohio, Pennsylvania, New Mexico, etc., or whatever state that's in the swing more important than the rest of the US? With a proportional system, every vote counts. With a district system, all regions count. Why is it good that a republican minority in New York doesn't get any representation, the same goes for a democratic minority in Idaho? Winner takes all seems decisive, but it's not all that fair, especially not on the national level. Maybe take a look at a true modern democracy, the democracy of Germany (invented by the US), and see how both regional and national interests are balanced in the Bundestag. You might learn from yourselves someday...
Given the normal distribution, if they're 95% sure it's within 3% of the stated number (thus plus or minus two (1.96) standard deviations), they are 99% sure that it's within 4.2% of the stated number. To get a difference larger than 5.6% this would be 1 in a 1000 (0.999). 8% would be once in a million. In jargon: normal tails die of quickly.
An inexact science, ...
More accurately, statistics is the science of the inexact, and actually a branch of mathematics. Any experiments done in science has statistical results, inexact results, where the inexactness is quantified. Statistics is about two main things: bias and variance. Bias occurs when either a theory is wrong, or the respondents in the theory, be it electrons or voters, refuse to give their readings. If it's done in a systematic way, it's called bias. Variance is a general uncertainty around this bias. Bias can be quantified, for instance that conservatives are less likely to respond to the poll. This can, and will, be catered for. Variance caters for the scaling up: if you queried 500 people, and try to scale up to the entire population, you can quantify this by taking the number of respondents into account. If both the bias and the variance don't match up with an election result, i.e., the calculation show that the actual outcome is improbable both with the bias (i.e., the expected outcome, catering for systematic error) and the variance (the uncertainty around this estimate), one of two things can be the matter: your calculation of bias is wrong, or the system is rigged. Blaming it purely on the statistics is very easy, but not necessarily correct. By publishing and researching the bias calculations you might be able to figure out which of the two is the case. Such a thing has not been done, or, at least, not been published.
In the end this leads to a whole lot of unanswered questions, meaning that the results are, and will be, in doubt. Taking care of this doubt should be a number one issue in running any new election, trying to reason it away is not helpful. In most countries it is accepted that the vote will be off by quite a bit, only in the US this is the difference between having one administration or a different one. Please observe Brazil at this moment. They are going into a second round because the incumbent didn't get a decisive vote. In France, the difference needs to be 5 percent. This takes care of a lot of error due to variance (due to unintentionally making the wrong vote), and the counting process. My opinion on this is that the oldest modern democracy in the world is in a crisis and needs to re-examine its process.
"I think therefore I am" is a statement of observation and reasoning, a very famous one. Even if the observation 'I think' was an assumption (which it isn't, how else can you observe something by thinking it. Think about this), this would not be faith, but belief: two totally different things. I am an atheist, I do not have faith. I however believe many things, including that the sun comes up each morning. This is not faith, this is a belief (and a bloody well founded one as well). Disentangling the mess that religion has made of the terms faith and belief is a step further towards understanding.
Reason without faith is perfectly reasonable. Reason with faith needs a bit of explanation.
But, in any case, I do live in a city, and I am forced to inhale car fumes of people that enjoy their mobility for no good reason. What are your thoughts on this, oh man with the swinging fist (*). You might enjoy your meal without being forced to inhale cigarette smoke, I on the other hand cannot enjoy a walk without the great stimulant of car fumes. Many studies have shown that kids growing up in cities have more health problems than kids in the country. Why is your mobility more important than my health?
(*) I'm actually six foot three, 210 pounds. I would love you taking a swing at me ;)
Where's faith in this equation?
Yet another fix
Keep your car out of my town, I would like to add.
Smoker and publicly transported.
Try to do the math and see where it breaks down. Vocabulary of students: 10,000 words (pretty low), number of possible 10 word sentences (short paragraphs): 10,000^10=1e40. Percentage of nonsense sentences due to grammaticality constraints: 0.99999 (wild guess, but add extra 9's if you so desire. Also consider level of knowledge of grammar by high-school students.). Leaves 1e35 possible paragraphs. Even if I'm 20 orders of magnitude off, the 22 million papers do not even make a dent in this space.
This is quite a wild guess, but we can also tackle it emperically. Take a random website with some text on it: take 10 consecutive words of that website and feed that to google (using quotes to make sure that you only get exact matches). Now count the results. You will either find your original back, or another website that has plagiarized your original (or vice-versa).
Just as a test, I took a message from the kde-man viewer: "Be careful that you must take care about upper case and lower case characters". Pretty innocuous right? Somebody might have made this particular grammatical construct before. Wrong. It's unique. Check the results.
Now try to write your own sentences of 10 words and see if you can get a hit. For instance: "Make sure that you only use upper and lower case characters". No, no hits. "Does rewriting a paper you found in your own words also count?" No hits. Maybe take your entire post and feed it, 10 consecutive words at a time, to Google. I'd be suprised if you get a single hit, given that your post hasn't been cached yet. You are truly a unique person that produces unique writing, even 10 words at a time!
In all honesty, I think the schools would be better off to put all old papers on a search engine-accessible part of their own website, and do an automatic check using whatever search engine that does the job (read: Google) for plagiarism. That way, there's no commercial entity involved that swallows student's work, and if many schools do this, they get the same network effect as with turnitin.
Hmm, I can think of four. And no Tomcat to be seen.
You're absolutely right: drug use has victims. However, the prisons and cops have probably destroyed more families (dad in prison for smoking), than they have saved. At a huge cost to society. And then there's still alcohol for the law-abiding citizen: now that's a home-wrecker exceeding most other things. Let alone the Colombian farmer whose livelihood gets destroyed to officially save these poor Americans from themselves as if these can't be trusted to lead their lives the way they want to. All very pathetic, really.
And what evidence gives you this ridiculous notion that giving up privacy ensures that your children are safer? What basis has your belief that your government is only interested in your best interests and will not misuse the freedoms you give them?
So on the one hand you have the unproven assertion that increased spying will actually gain security (*), while on the other hand you have the historical certainty that powers given to the government will be abused to the full extent of the law (**). Your children will be proud of you.
(*) Note that there was sufficient intelligence to prevent 9-11. It was just not used.
(**) And already librarians, Quakers and knitting clubs are being harassed.
Given that this particular State's officials might actually be in power exactly due to the interferebility of the Diebold machines, why do you think they'll endanger their reelection by throwing out Diebold?
(*)yes, the second lemma of many dictionaries in the US do state that a republic is a form of representative democracy, yet the first lemma is invariably a 'country that has a president (elected or not) as head of state'. Indeed this does seem to sum up the breadth of democracy the US is experiencing now, but I don't think that's what you meant.
Or could it be that the Dem county Dem vote was so much lower than expected that they tried to fix that one first, given that any irregularities would have been tried there?
Luckily SCOTUS decided that it was way more important to have a president before the 15th of January than to figure out who should be Prez. Can't blame them, any figurehead will do: it's following the rituals of Democracy that make a country Democratic.
Computer monocultures are inherently unstable and when corrupted once, the mono-culture can be brought down exponentially fast. This holds for any mono-culture in computing: security software not excluded.
If I want any office-type of job, whether computer-related or simply sales, I'd better know how to operate windows/office, otherwise: no job. If I want to be able to work from home from time to time, I'd better have at least one windows/office machine at home, whether company paid or by myself. If I choose to use OpenOffice, I'd better be humble when something goes wrong in cooperating. Although it is accepted that office fucks up itself from time to time, it's not accepted of any other app. By the job. The big monopoly of Microsoft is in the workplace, not at the homes.
The choice for or against MS is simple for many: job or no job. Is that enough of a monopoly for you?
Probably Google send the newspapers the instructions on how to prevent being cached, and considered the case done. If they insist shooting themselves in the foot, let it be.
hmm, reading further in the thread, apparently my lack of knowledge of the syntax of robots.txt is only matched by my willingness to make a fool of myself. If indeed NOCACHE is an option, and respected by Google, then these Belgians are stupid to go to court for something that has an easy technological fix and the court is stupid to allow this.
Can you use robots.txt for this? The situation is this: I have an article in /public/article.html, and later move it into my archive at /archive/article.html, where only paid subscribers can see it. My robots.txt dutifully notices that /archive is off-limits. How to tell Google to delete their cached version of article.html?
Apparently only by making /public offlimits as well. This is not the intention.