It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked."
If you were talking about one or a few polls I would agree with your premise that we should not investigate these allegations. However, it is valid to notice that every single poll agency in the United States, including the agency hired by Trump, they all failed to even come closed to their statistical predictions.
Nope. The analysis by 538 published before the election was that Clinton was ahead by a number that was equal to the polling margin of error. "it shouldn’t be hard to see how Clinton could lose. She’s up by about 3 percentage points nationally, and 3-point polling errors happen fairly often, including in the last two federal elections." (November 6: http://fivethirtyeight.com/fea... )
As Entrope points out, "statistical" adjustments are a very bad idea. Since the actual votes are not paired to the voters, that would mean "adjusting" (i.e., throwing out votes) based on how (some authority) decides to guess the illegal voters voted. Analyzing voter rolls to identify illegal voters would be a good idea--- if nothing else, as a way to try to kill that idiotic statement repeated over and over again by right-wing nuts with no evidence whatsoever that the election is invalid because so many illegal people vote. Although the problem with trying to kill conspiracy theories with data is that the people who believe the conspiracy theories just dismiss the data as more evidence for how widespread the conspiracy actually is.
So all those people across the country destroying property are not Democrats? Nor are all the people signing petitions to turn electors against Trump?
This is an example of what should be on the list of logical fallacies, the "ambiguous specifier". The post said "Why won't Democrats support the outcome?" What does that word "Democrats" specify? All Democrats? Some Democrats? Most Democrats? At least one Democrat? The official Democratic policy? The Democratic leadership?
The Democratic Party stated that they accepted the outcome of the election clearly, explicitly, and without qualifications. The Democrats most explicitly did support the outcome. If some individuals apparently don't, they are not doing so with the support of the Democratic party.
You DO NOT want the company you are buying things from to make record profits.
Standard economic theory says that they will make record profits if and only if their customers believe that they are providing a superior product. So, yes, actually you do want to buy from a company making record profits: this is a sign that their customers like them; and, in this case, iPhones have been around long enough that it is a sign that their customers are repeat customers.
Actually, America's interests would be served by doing a recount of some portio of the ballots to verify accuracy. Quite apart from who won, it's valuable to check, check, and check again to verify if there is an error or tampering.
But, yes, it may not be in the Democratic Party's best interest. Although to be frank, they are already being labelled "sore losers" despite conceding the election and explicitly instructing their supporters to accept the results, so I doubt it would make any difference in how they are perceived.
The p-value you "calculate" is not for the hypothesis "the election was hacked." It is for the hypothesis "counties with electronic-only voting machines vote differently than counties with paper-trail voting machines." One, but only one, explanation for why they might be different is that the electronic, but not the paper trail, voting machines were hacked. The other explanation, not ruled out, is that the type of voting machine is indicative of counties that are different in other ways as well.
Also, I note that you are "computing" p-values without actually looking at data-- basically, you're recycling rumors. What is the standard deviation by county for counties that have electronic-only voting, and what is the deviation for counties that don't? You don't have that data. So, you actually can't calculate statistics.
The article in question here is from a computer security specialist who is not with the campaign or with the Democratic party saying that the Democratic party should ask for a recount
It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked." What he suggests is that it would be valuable to do the testing to verify: to "help allay doubt and give voters justified confidence that the results are accurate."
From this point of view, it does make sense: "trust but verify". It also makes sense to do something he doesn't suggest, which is to break down some of the electronic voting machines and inspect the code for malware (he only suggests comparing the paper trail to the electronic count, not looking at the machines that don't have a paper trail.)
Does anyone have a prediction of "bad stuff" made in an IPCC report that has actually happened?... The old reports are now old enough that there predictions should be apparent by now. I have reviewed the reports, and the cases I looked at (sea level rise, crop failures, fishing) were all falsified by what happened in reality.
You state that predictions are "falsified by what happened in reality", but you failed to show evidence for that. I am actually quite interested. Can you point to a specific published prediction from the IPCC-- one with a date that is not still in the future-- and show me data saying that the actual result was different from the prediction by more than the published error bars?
Of course, one of the main reasons WHY there had been high levels of ice in the antarctic is BECAUSE of the reduced ice in the arctic.
This is almost certainly not true. They are literally poles apart, and while, figuratively, it's a small world it's actually a pretty damn big planet.
Despite being "poles apart", it turns out that there is some amount of anticorrelation between the Arctic and Antarctic temperature variations (the "polar see-saw"-- sometimes called the "bipolar see-saw"). So, while they are "poles apart", they are still in the same system.
It is true, though, that the salinity of water in the circum-Antarctic ocean really doesn't much depend on Arctic melting. It is significantly impacted by Antarctic melting, though, with the odd result that melting glacial ice actually can increase the seasonal ice. Which is exactly what you point out further in your post:
In fact, it can take hundreds to thousands of years for water from the poles to reach the equator, and much of the water that does 'reach' the equator tends to be turned back towards the pole from which it came, due to equatorial up-welling and circulatory currents. Note: the above is a bit of a generalisation and a vast simplification to make the point. Feel free to investigate the matter further, I'm just trying to demonstrate the error inherent in parent's post.
The ice-melt makes the oceans fresher and fresher water freeze more easily than salty water.
This is true however, and is likely to be part of the reason why the area of antarctic sea ice has been quite high. Antarctica is a land mass, covered in ice. Some of that ice has been observed melting, and some actually sliding into the ocean. It is this 'land ice' becoming 'sea ice' either directly, ice shelves sliding into the ocean, or indirectly, melting and refreezing, that accounts for the observations.
But how do you arrive at the "lowest feasible number" ? Why is 30 feasible? What if only 300 is feasible?
Good question. Let's do some math. From a year by year temperature graph http://data.giss.nasa.gov/gist... we see random temperation variation is somewhere about 0.3C. Temperature rise is on the order of 0.15 degrees per decade. Averaging reduces the noise by the square root of the number of points (Poisson statistics*). So, the temperature rise (signal) is larger then the noise (year to year variations) when 0.015*N> 0.3/SQRT(N). Thus, N^(3/2) = 0.3/0.015, and we calculate N = 7.3 years.
So, in 7.3 years the signal (temperature rise) is roughly equal to the statistical noise (year to year variation). Science typically likes to not draw conclusions until you get at least 3 standard deviations, so that would be about 20 years.
--- *footnote: correctly, Poisson statistics are dependent on the number of independent points. Year to year temperatures, however, are not completely independent-- they show some amount of correlation ("autocorrelation"). So the number of points should actually be reduced by the aurocorrelation coefficient. That will bump the number of points N up slightly. So, actually, 30 years is probably a pretty good number to guess.
A politician would simply never be willing to destroy his career in politics in order to do what he believes is right. If it means not getting another term, it won't get done.
Lyndon Johnson did. Reputedly, when he signed the Civil Rights Act, he commented "we have lost the south (for the Democratic party) for a generation."
His "generation" turned out to be a long one; that was 1964, and there is no sign of the south voting for Democrats again any time soon.
Meh. The real situation is what it is, and that seems to be plenty to discuss; I don't see the point in fantasizing catastrophe scenarios that have very little real science behind them. For a look at what the best actual expectations are for the impact of warming and loss of sea ice, the WG-II report is still the best review: http://www.ipcc.ch/report/ar5/... (that's rather long, but the 32 page summary is here: http://ipcc-wg2.gov/AR5/images... )
Somebody is exaggerating... Even the mistake was rounded up: (9/5)*35 = 19.444
First, your conversion factor is upside down. 5/9, not 9/5.
Second, you have it backwards-- the number from the original source was 20 Celsius, and converted into Fahrenheit for the popular article. 20*9/5 = 36F, which was rounded DOWN to 35. (Correctly, since the original number was not written to two figure precision).
They can, but it would be a PR nightmare and would be splattered all over the news within a day of it happening.
No.
It appears that when this kind of thing happens it's "splattered all over the news", but that's because when it doesn't make the news, you never hear about it. It's a perception error.
Your statement that the mainstream media didn't cover the story is wrong. The mainstream media, including the usual "liberal" press such as the New York Times and the Washington Post, were all over the story that the documents 60 Minutes had didn't look authentic.
60 Minutes retracted it. They retracted it in public, in prime time, and fired the people responsible.
Making mistakes happens. People screw up. What matters is whether they correct it. And they did.
Gosh, it's sweet that the Breitbart makes a retraction... when it's misattributed quote from somebody who was politically on their side. Too bad they are a little lax on other errors. They figure if it's kinda similar to real news, that's good enough.
You want to preserve democratic freedoms by censoring fake news? Wouldn't that be like... taking away the most important democratic freedom of them all, free speech?
No.
Recognizing that a problem exists is not the same as proposing censorship as a solution. I read the article-- apparently you didn't-- and at no place does he suggest censorship as the solution. You are the one who just proposed censorship.
The problem exists. If even talking about the problem draws cries from the likes of you of "you're proposing censorship"-- this is just about as bad in terms of squashing speech. The data shows that fake news is getting more clicks than real news (the link in the article you didn't read: https://www.buzzfeed.com/craig... )
We have a problem. What's the approach to a solution? You are the one who just proposed censorship. Have any other ideas?
I mean, it's really odd. Everybody who's so fucking concerned about fake news now didn't seem to be too upset when Dan Rather was pushing fake news.
A lot of people were upset. The mainstream media-- the same usual suspects you call left-leaning-- came down on 60 Minutes like a ton of bricks. 60 Minutes ran a correction in prime time, fired the people involved in approving the story to go on the air with inadequate fact checking, and asked Dan Rather to resign.
That seems pretty "upset" to me.
So, your demonstration that mainstream media does fake news is a single news story fourteen years ago that was based on documents that turned out to be forged, for which the staff of the program was fired or asked to resign.
Just think what CBS' Dan Rather did with the Bush reports years ago. They made them up out of thin air.
Yes, Dan Rather of 60 Minutes did report a story based on documents that were turned out to be fake. And, guess what? A day and a half later, The Washington Post, The New York Times. USA Today and the Associated Press all ran stories disputing the documents. If you're tallking about the mainstream media running fake stories-- how about the mainstream media reporting that fake news was fake. 60 Minutes eventually ran a public retraction of the story, and subsequently fired the people responsible.
This is the mainstream media doing self-correction. That was back in those halcyon days of yesterday when integrity in journalism was actually an ideal that the media tried for, before all the internet media started going with the plan "who cares it it's true? If it gets clicks, run with it!"
Do the fake news sites ever run retractions? Has Breitbart ever retracted anything they've said?
Slashdot Mirror
I highly doubt the Democrat party would want to risk verifying accuracy since it is more likely that Democrats commit voter fraud then Republicans.
No evidence whatsoever for that statement.
It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked."
If you were talking about one or a few polls I would agree with your premise that we should not investigate these allegations. However, it is valid to notice that every single poll agency in the United States, including the agency hired by Trump, they all failed to even come closed to their statistical predictions.
Nope. The analysis by 538 published before the election was that Clinton was ahead by a number that was equal to the polling margin of error. "it shouldn’t be hard to see how Clinton could lose. She’s up by about 3 percentage points nationally, and 3-point polling errors happen fairly often, including in the last two federal elections." (November 6: http://fivethirtyeight.com/fea... )
Another one worth reading: http://fivethirtyeight.com/fea...
As Entrope points out, "statistical" adjustments are a very bad idea. Since the actual votes are not paired to the voters, that would mean "adjusting" (i.e., throwing out votes) based on how (some authority) decides to guess the illegal voters voted.
Analyzing voter rolls to identify illegal voters would be a good idea--- if nothing else, as a way to try to kill that idiotic statement repeated over and over again by right-wing nuts with no evidence whatsoever that the election is invalid because so many illegal people vote.
Although the problem with trying to kill conspiracy theories with data is that the people who believe the conspiracy theories just dismiss the data as more evidence for how widespread the conspiracy actually is.
they are already being labelled "sore losers" despite conceding the election and explicitly instructing their supporters to accept the results
They're not being "labeled" as "sore losers". They are if fact being sore losers. Challenging the results of an election is not "conceding".
The whole point of TFA is that the Democrats are not challenging the result of the election, but the writer thinks that they should.
Learn to read.
Apparently you don't know what a "recount" is.
Once a vote is recorded, the information on who cast that vote is not retained. "Recount" won't tell you which votes were made by which people.
So all those people across the country destroying property are not Democrats? Nor are all the people signing petitions to turn electors against Trump?
This is an example of what should be on the list of logical fallacies, the "ambiguous specifier". The post said "Why won't Democrats support the outcome?" What does that word "Democrats" specify? All Democrats? Some Democrats? Most Democrats? At least one Democrat? The official Democratic policy? The Democratic leadership?
The Democratic Party stated that they accepted the outcome of the election clearly, explicitly, and without qualifications. The Democrats most explicitly did support the outcome. If some individuals apparently don't, they are not doing so with the support of the Democratic party.
You DO NOT want the company you are buying things from to make record profits.
Standard economic theory says that they will make record profits if and only if their customers believe that they are providing a superior product. So, yes, actually you do want to buy from a company making record profits: this is a sign that their customers like them; and, in this case, iPhones have been around long enough that it is a sign that their customers are repeat customers.
Actually, America's interests would be served by doing a recount of some portio of the ballots to verify accuracy. Quite apart from who won, it's valuable to check, check, and check again to verify if there is an error or tampering.
But, yes, it may not be in the Democratic Party's best interest. Although to be frank, they are already being labelled "sore losers" despite conceding the election and explicitly instructing their supporters to accept the results, so I doubt it would make any difference in how they are perceived.
Nope.
The p-value you "calculate" is not for the hypothesis "the election was hacked." It is for the hypothesis "counties with electronic-only voting machines vote differently than counties with paper-trail voting machines." One, but only one, explanation for why they might be different is that the electronic, but not the paper trail, voting machines were hacked. The other explanation, not ruled out, is that the type of voting machine is indicative of counties that are different in other ways as well.
Also, I note that you are "computing" p-values without actually looking at data-- basically, you're recycling rumors. What is the standard deviation by county for counties that have electronic-only voting, and what is the deviation for counties that don't? You don't have that data. So, you actually can't calculate statistics.
"Why won't Democrats support the outcome?"
You do realize that the Democrats do support the outcome, right? Hillary Clinton conceded and called on her supporters to accept the US election result. Hillary's campaign is not contesting the election.
The article in question here is from a computer security specialist who is not with the campaign or with the Democratic party saying that the Democratic party should ask for a recount
It's worth noting that if you actually read the article, he doesn't say that the ballots actually were hacked: in fact, what he says is "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked." What he suggests is that it would be valuable to do the testing to verify: to "help allay doubt and give voters justified confidence that the results are accurate."
From this point of view, it does make sense: "trust but verify". It also makes sense to do something he doesn't suggest, which is to break down some of the electronic voting machines and inspect the code for malware (he only suggests comparing the paper trail to the electronic count, not looking at the machines that don't have a paper trail.)
Does anyone have a prediction of "bad stuff" made in an IPCC report that has actually happened? ... The old reports are now old enough that there predictions should be apparent by now. I have reviewed the reports, and the cases I looked at (sea level rise, crop failures, fishing) were all falsified by what happened in reality.
You state that predictions are "falsified by what happened in reality", but you failed to show evidence for that.
I am actually quite interested. Can you point to a specific published prediction from the IPCC-- one with a date that is not still in the future-- and show me data saying that the actual result was different from the prediction by more than the published error bars?
Nice links, by the way-- very informative.
Of course, one of the main reasons WHY there had been high levels of ice in the antarctic is BECAUSE of the reduced ice in the arctic.
This is almost certainly not true. They are literally poles apart, and while, figuratively, it's a small world it's actually a pretty damn big planet.
Despite being "poles apart", it turns out that there is some amount of anticorrelation between the Arctic and Antarctic temperature variations (the "polar see-saw"-- sometimes called the "bipolar see-saw"). So, while they are "poles apart", they are still in the same system.
It is true, though, that the salinity of water in the circum-Antarctic ocean really doesn't much depend on Arctic melting. It is significantly impacted by Antarctic melting, though, with the odd result that melting glacial ice actually can increase the seasonal ice. Which is exactly what you point out further in your post:
In fact, it can take hundreds to thousands of years for water from the poles to reach the equator, and much of the water that does 'reach' the equator tends to be turned back towards the pole from which it came, due to equatorial up-welling and circulatory currents. Note: the above is a bit of a generalisation and a vast simplification to make the point. Feel free to investigate the matter further, I'm just trying to demonstrate the error inherent in parent's post.
The ice-melt makes the oceans fresher and fresher water freeze more easily than salty water.
This is true however, and is likely to be part of the reason why the area of antarctic sea ice has been quite high. Antarctica is a land mass, covered in ice. Some of that ice has been observed melting, and some actually sliding into the ocean. It is this 'land ice' becoming 'sea ice' either directly, ice shelves sliding into the ocean, or indirectly, melting and refreezing, that accounts for the observations.
...
But how do you arrive at the "lowest feasible number" ? Why is 30 feasible? What if only 300 is feasible?
Good question. Let's do some math.
From a year by year temperature graph http://data.giss.nasa.gov/gist... we see random temperation variation is somewhere about 0.3C. Temperature rise is on the order of 0.15 degrees per decade. Averaging reduces the noise by the square root of the number of points (Poisson statistics*). So, the temperature rise (signal) is larger then the noise (year to year variations) when 0.015*N> 0.3/SQRT(N). Thus, N^(3/2) = 0.3/0.015, and we calculate N = 7.3 years.
So, in 7.3 years the signal (temperature rise) is roughly equal to the statistical noise (year to year variation). Science typically likes to not draw conclusions until you get at least 3 standard deviations, so that would be about 20 years.
---
*footnote: correctly, Poisson statistics are dependent on the number of independent points. Year to year temperatures, however, are not completely independent-- they show some amount of correlation ("autocorrelation"). So the number of points should actually be reduced by the aurocorrelation coefficient. That will bump the number of points N up slightly. So, actually, 30 years is probably a pretty good number to guess.
A politician would simply never be willing to destroy his career in politics in order to do what he believes is right. If it means not getting another term, it won't get done.
Lyndon Johnson did. Reputedly, when he signed the Civil Rights Act, he commented "we have lost the south (for the Democratic party) for a generation."
His "generation" turned out to be a long one; that was 1964, and there is no sign of the south voting for Democrats again any time soon.
https://sites.google.com/site/...
https://en.wikipedia.org/wiki/...
Meh.
The real situation is what it is, and that seems to be plenty to discuss; I don't see the point in fantasizing catastrophe scenarios that have very little real science behind them.
For a look at what the best actual expectations are for the impact of warming and loss of sea ice, the WG-II report is still the best review: http://www.ipcc.ch/report/ar5/...
(that's rather long, but the 32 page summary is here: http://ipcc-wg2.gov/AR5/images... )
Somebody is exaggerating... Even the mistake was rounded up: (9/5)*35 = 19.444
First, your conversion factor is upside down. 5/9, not 9/5.
Second, you have it backwards-- the number from the original source was 20 Celsius, and converted into Fahrenheit for the popular article. 20*9/5 = 36F, which was rounded DOWN to 35. (Correctly, since the original number was not written to two figure precision).
They can, but it would be a PR nightmare and would be splattered all over the news within a day of it happening.
No.
It appears that when this kind of thing happens it's "splattered all over the news", but that's because when it doesn't make the news, you never hear about it. It's a perception error.
After that... use a gmail account which you also pull down via IMAP to a local server ...
The discussion here is about people losing access to their gmail account without notice and with no way to get it back.
The whole POINT of the internet is to share cute cat photos. Clearly, Zuckerberg is doing it wrong!
http://gizmodo.com/why-cats-ru...
So, you agree the MSM jumped on it.
So, the mainstream media correction works.
Your statement that the mainstream media didn't cover the story is wrong. The mainstream media, including the usual "liberal" press such as the New York Times and the Washington Post, were all over the story that the documents 60 Minutes had didn't look authentic.
60 Minutes retracted it. They retracted it in public, in prime time, and fired the people responsible.
Making mistakes happens. People screw up. What matters is whether they correct it. And they did.
Gosh, it's sweet that the Breitbart makes a retraction... when it's misattributed quote from somebody who was politically on their side. Too bad they are a little lax on other errors. They figure if it's kinda similar to real news, that's good enough.
You want to preserve democratic freedoms by censoring fake news? Wouldn't that be like... taking away the most important democratic freedom of them all, free speech?
No.
Recognizing that a problem exists is not the same as proposing censorship as a solution. I read the article-- apparently you didn't-- and at no place does he suggest censorship as the solution. You are the one who just proposed censorship.
The problem exists. If even talking about the problem draws cries from the likes of you of "you're proposing censorship"-- this is just about as bad in terms of squashing speech. The data shows that fake news is getting more clicks than real news (the link in the article you didn't read: https://www.buzzfeed.com/craig... )
We have a problem. What's the approach to a solution? You are the one who just proposed censorship. Have any other ideas?
I mean, it's really odd. Everybody who's so fucking concerned about fake news now didn't seem to be too upset when Dan Rather was pushing fake news.
A lot of people were upset. The mainstream media-- the same usual suspects you call left-leaning-- came down on 60 Minutes like a ton of bricks. 60 Minutes ran a correction in prime time, fired the people involved in approving the story to go on the air with inadequate fact checking, and asked Dan Rather to resign.
That seems pretty "upset" to me.
So, your demonstration that mainstream media does fake news is a single news story fourteen years ago that was based on documents that turned out to be forged, for which the staff of the program was fired or asked to resign.
Just think what CBS' Dan Rather did with the Bush reports years ago. They made them up out of thin air.
Yes, Dan Rather of 60 Minutes did report a story based on documents that were turned out to be fake. And, guess what? A day and a half later, The Washington Post, The New York Times. USA Today and the Associated Press all ran stories disputing the documents. If you're tallking about the mainstream media running fake stories-- how about the mainstream media reporting that fake news was fake. 60 Minutes eventually ran a public retraction of the story, and subsequently fired the people responsible.
This is the mainstream media doing self-correction. That was back in those halcyon days of yesterday when integrity in journalism was actually an ideal that the media tried for, before all the internet media started going with the plan "who cares it it's true? If it gets clicks, run with it!"
Do the fake news sites ever run retractions? Has Breitbart ever retracted anything they've said?