(sorry, they DO NOT despite what anyone has told you; for example, an iCal calendar item's title won't show up properly in Mozilla Calendar.)
I never said they did. I'm just saying they're revamping the product to handle multiple handhelds. I should have also mentioned that they want to be able to work with other calendars as well. I assume that Sunbird 0.2 has many bugs, which they are aware of, and hope to fix.
It's pretty astounding that a simple file format like a frigging CALENDAR can't be standardized across calendar programs which all claim to be able to use the same...uh...standard file format.
It must be hard, isn't it? To be so much smarter than the rest of the world, to know exactly how Calendar programs should work. But sadly, nobody else has the wisdom and insight you do...
Seriously, I'm done with you. I can't believe someone bothered replying to a simple informative post with yelling and ranting.
Sunbird is currently undergoing a complete rewrite. I've worked with early builds of the new Sunbird, and it's looking pretty nice. Eventually, it should emerge with a much better framework to handle many Calendar and scheduling needs.
Mozilla Lightning is also doing well in development. You can see some screenshots of it here (may load slowly): http://diary.e-gandalf.net/?p=35.
It seems like these developers finally understand the great need for Calendar products. I frequently hear discussion of the most wanted features, such as different calendar formats, integration with other handhelds, etc.
It does not matter if the government has actually abused citizens via the Patriot Act. The only thing that matters is that it can.
Except that the vast majority of laws can be abused.
But you may say "But the Patrioct Act has a much greater chance for abuse!" That's an excellent point, one based on theory. But this paper would prove otherwise, that past history has shown the Patriot Act may indeed be abused rarely, at about the same rate that other laws are abused.
If so, then by your logic, we should get rid of the vast majority of laws, as they could be abused as often as the Patriot Act is.
Fifty-seven years later Americans are demanding their pound of flesh from the French with the incumbent George Bush's "You are either with us or against us." Other princes of darkness - and there are many in America -- are crediting the President of France with dozens of qualities and the French with more character traits, all negative.
So you countered French leader bashing with America leader bashing.
That we shoulnd't trust papers that haven't gone through peer process. The researchers went in front of the media and acted like their work is solid. And the media relayed that. I wanted to make sure people understood these results are very preliminary, and should be viewed with caution.
Looking back, I probably should have worded it differently, since I didn't realize there really isn't a fast track process to publish results. So I should have probably worded it differently, saying these results are preliminary and haven't been given a standard thumbs up by other statisticians.
The researchers didn't do anything wrong here. They made it clear that this was just a working paper. It is the reporters who are not stating clearly enough that these results are preliminary
No they didn't, they made it clear they thought they were already right...so right in fact...they could go in front of the media now instead of waiting for proper peer review to make an extraordinary claim.
"No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," he said in a statement. "There is just a trivial probability of evidence like this appearing in a population where the true difference is zero -- less than once in a thousand chances."
That doesn't sound like they thought it was preliminary to me. And that's the problem...they are potentially misleading millions because they skipped the peer review process.
I'm telling you not to believe preliminary stastical reports that obviously have not gone through proper procedures. That's it. The rest is my opinion why I think the paper is wrong. All you have is a simple regression study using a lot of assumptions and few variables.
To make this group even more questionable, they go in front of the media and act as if their claims of some stastical problem are there. "No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," he said in a statement. "There is just a trivial probability of evidence like this appearing in a population where the true difference is zero -- less than once in a thousand chances."
No matter how many factors or variables? They didn't even factor in Nader!
It's a working paper (and advertised as such). The data are available for replication. This is standard pratice.
But it is not standard practice to work with the media so closely when the results are preliminary and haven't been vetted by other professionals. This is wrong, as it muddies up the waters for those who don't realize this is a preliminary paper, and not a conclusion.
And...I need to better explain what I previously said so it makes sense. They are trying to find a regression line between electronic voting counties and non eletronic voting counties, using percent change of voters for a candidate between 2000 and 2004, and if the lines are different, they found a problem.
My problem is the regression lines they used and the whole manner they used to determine differences between areas. Like I said, they ignored Nader's effect completely, and he can easily change the results in heavily Democratic areas. Anyways, I don't think the regression techniques they used are proper, and ultimately, I don't believe regression lines can be used to really explain anything in manner they are attempting to. They are assuming a consistant trend between counties, so they picked regression lines to fit their assumptions. But that could very easily be a bad assumption, since Nader's influence could very easily be very different in urban areas than in rural areas...but unfortunately, they ignored Nader completely.
What I got out of it is that the light purple line represents the electronic voting trends, while the darker line represents on electronic voting trends.
Now, as you can see, neither line fits the data points too well on the far right of the graph, which means (in my opinion), that they picked a horrible regression technique. The regression lines fit well in the middle of the graph, but not on the right. And that's where they are making their claims at, on the right part of the graph. They are trying to extrapolate data off of a bad regression line and using that as proof of a problem.
Also, the paper's lead author is a full professor, not a student.
Aaah, you're right. Three sociology students and a sociology professor. I would assume that they aren't very statistically minded. I make that assumption based off of a humorous part of their paper. They weighted (?) variables in a regression analysis, and called the process "complicated".
but it's common practice --- in all fields --- to circulate working papers and drafts.
But it is not common to rush to the media to announce them as they are doing.
"too many of these wannabe statiticians are not publishing their results."
Published is a term often used to describe a paper that has been approved and printed in a respected journal. This paper was not published in that sense.
Well, if you wanted to take it one step further, you could weigh the variables!
Now before you say "Whoa! That's crazy talk! Weighing variables to take population size into account? That's too hard!" Don't worry, our Berkeley students who are masters of statistics, pulled it off! It wasn't easy though, the said it was a "complicated" procedure...
But isn't it curious that the largest disparity between expected and actual e-voting results occurred in heavily democratic counties?
Isn't it interesting that you instantly believed a report written by four students? That this report was not published by any respected journals? That they did not even try to go through a respectiable scientific process before rushing to the media? That the only ones who runthis story are internet blogs and PRNewswire? That their study didn't even include Nader as a variable? That they call simple statistical techniques "complicated"?
I would like to point out one extremely important fact. This paper was not published in a journal. This was a simple statistical analysis by four students.
Let the scientific method work this out. If a paper has merit, let it be analyzed by stastic professionals, and if it does have merit, any statistical journal would be happy to run a major news story that would give them publicitiy.
But too many of these wannabe statiticians are not publishing their results. They make unrelaistic assumptions, they use questionable approaches to making claims, they don't use enough variables (in the case of this report, they didn't even factor in Nader!) And when they find something they believe is significant, they bypass the scientific method completely, and rush straight to internet blogs or PRNewswire.
Again, let the scientific method take its course, and be very cautious of anything that doesn't.
Slashdot Mirror
The ACLU's argument against this law fails to mention that filtering can only be done on request of the customer.
Now why would the ACLU leave out that most important detail?
Whoa, calm down there.
(sorry, they DO NOT despite what anyone has told you; for example, an iCal calendar item's title won't show up properly in Mozilla Calendar.)
I never said they did. I'm just saying they're revamping the product to handle multiple handhelds. I should have also mentioned that they want to be able to work with other calendars as well. I assume that Sunbird 0.2 has many bugs, which they are aware of, and hope to fix.
It's pretty astounding that a simple file format like a frigging CALENDAR can't be standardized across calendar programs which all claim to be able to use the same...uh...standard file format.
It must be hard, isn't it? To be so much smarter than the rest of the world, to know exactly how Calendar programs should work. But sadly, nobody else has the wisdom and insight you do...
Seriously, I'm done with you. I can't believe someone bothered replying to a simple informative post with yelling and ranting.
Mozilla Lightning is also doing well in development. You can see some screenshots of it here (may load slowly): http://diary.e-gandalf.net/?p=35.
It seems like these developers finally understand the great need for Calendar products. I frequently hear discussion of the most wanted features, such as different calendar formats, integration with other handhelds, etc.
You have to agree this legislation is draconian and excessive, to say the least."
No I don't agree with you.
If it doesn't let me custom build my own radio station, then there is no way I'm switching from LaunchCast from Yahoo
Its cheaper too. =)
It does not matter if the government has actually abused citizens via the Patriot Act. The only thing that matters is that it can. Except that the vast majority of laws can be abused.
But you may say "But the Patrioct Act has a much greater chance for abuse!" That's an excellent point, one based on theory. But this paper would prove otherwise, that past history has shown the Patriot Act may indeed be abused rarely, at about the same rate that other laws are abused.
If so, then by your logic, we should get rid of the vast majority of laws, as they could be abused as often as the Patriot Act is.
Why do you expect it to be?
I expect people to not be misleading. I don't care who they work for.
Hmm...starting your article with a misleading flamewar rant against a politician? It's right on par with Slashdot's level of professionalism.
Look, the chances are now 1 in 2!!!
http://neo.jpl.nasa.gov/cgi-bin/ip?0.5e-0
Actually, it's not. The URL just lets you plug in any scientific notation number. Hope I didn't alarm you too much. =)
Fifty-seven years later Americans are demanding their pound of flesh from the French with the incumbent George Bush's "You are either with us or against us." Other princes of darkness - and there are many in America -- are crediting the President of France with dozens of qualities and the French with more character traits, all negative.
So you countered French leader bashing with America leader bashing.
1 in 455 chance of humanity being wiped out in the next 100 years?
So every 45500 years, a mass extinction event takes place on Earth? That sure doesn't sound right.
I'm starting to think Jon Katz is submitting all these articles under pseudonyms.
No, its just michael. This is normal for him.
What is your point?
That we shoulnd't trust papers that haven't gone through peer process. The researchers went in front of the media and acted like their work is solid. And the media relayed that. I wanted to make sure people understood these results are very preliminary, and should be viewed with caution.
Looking back, I probably should have worded it differently, since I didn't realize there really isn't a fast track process to publish results. So I should have probably worded it differently, saying these results are preliminary and haven't been given a standard thumbs up by other statisticians.
The researchers didn't do anything wrong here. They made it clear that this was just a working paper. It is the reporters who are not stating clearly enough that these results are preliminary
No they didn't, they made it clear they thought they were already right...so right in fact...they could go in front of the media now instead of waiting for proper peer review to make an extraordinary claim.
"No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," he said in a statement. "There is just a trivial probability of evidence like this appearing in a population where the true difference is zero -- less than once in a thousand chances."
That doesn't sound like they thought it was preliminary to me. And that's the problem...they are potentially misleading millions because they skipped the peer review process.
I'm telling you not to believe preliminary stastical reports that obviously have not gone through proper procedures. That's it. The rest is my opinion why I think the paper is wrong. All you have is a simple regression study using a lot of assumptions and few variables.
To make this group even more questionable, they go in front of the media and act as if their claims of some stastical problem are there. "No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," he said in a statement. "There is just a trivial probability of evidence like this appearing in a population where the true difference is zero -- less than once in a thousand chances."
No matter how many factors or variables? They didn't even factor in Nader!
Here is one problem I have with their methodology.
It's a working paper (and advertised as such). The data are available for replication. This is standard pratice.
But it is not standard practice to work with the media so closely when the results are preliminary and haven't been vetted by other professionals. This is wrong, as it muddies up the waters for those who don't realize this is a preliminary paper, and not a conclusion.
Argh "on electronic" means "non electronic"
And...I need to better explain what I previously said so it makes sense. They are trying to find a regression line between electronic voting counties and non eletronic voting counties, using percent change of voters for a candidate between 2000 and 2004, and if the lines are different, they found a problem.
My problem is the regression lines they used and the whole manner they used to determine differences between areas. Like I said, they ignored Nader's effect completely, and he can easily change the results in heavily Democratic areas. Anyways, I don't think the regression techniques they used are proper, and ultimately, I don't believe regression lines can be used to really explain anything in manner they are attempting to. They are assuming a consistant trend between counties, so they picked regression lines to fit their assumptions. But that could very easily be a bad assumption, since Nader's influence could very easily be very different in urban areas than in rural areas...but unfortunately, they ignored Nader completely.
What I got out of it is that the light purple line represents the electronic voting trends, while the darker line represents on electronic voting trends.
Now, as you can see, neither line fits the data points too well on the far right of the graph, which means (in my opinion), that they picked a horrible regression technique. The regression lines fit well in the middle of the graph, but not on the right. And that's where they are making their claims at, on the right part of the graph. They are trying to extrapolate data off of a bad regression line and using that as proof of a problem.
Also, the paper's lead author is a full professor, not a student.
Aaah, you're right. Three sociology students and a sociology professor. I would assume that they aren't very statistically minded. I make that assumption based off of a humorous part of their paper. They weighted (?) variables in a regression analysis, and called the process "complicated".
but it's common practice --- in all fields --- to circulate working papers and drafts.
But it is not common to rush to the media to announce them as they are doing.
"too many of these wannabe statiticians are not publishing their results."
Published is a term often used to describe a paper that has been approved and printed in a respected journal. This paper was not published in that sense.
Well, if you wanted to take it one step further, you could weigh the variables!
Now before you say "Whoa! That's crazy talk! Weighing variables to take population size into account? That's too hard!" Don't worry, our Berkeley students who are masters of statistics, pulled it off! It wasn't easy though, the said it was a "complicated" procedure...
But isn't it curious that the largest disparity between expected and actual e-voting results occurred in heavily democratic counties?
Isn't it interesting that you instantly believed a report written by four students? That this report was not published by any respected journals? That they did not even try to go through a respectiable scientific process before rushing to the media? That the only ones who runthis story are internet blogs and PRNewswire? That their study didn't even include Nader as a variable? That they call simple statistical techniques "complicated"?
They also neglect Nader. He was a big variable between changes in 2000 and 2004 election results.
I would like to point out one extremely important fact. This paper was not published in a journal. This was a simple statistical analysis by four students.
Let the scientific method work this out. If a paper has merit, let it be analyzed by stastic professionals, and if it does have merit, any statistical journal would be happy to run a major news story that would give them publicitiy.
But too many of these wannabe statiticians are not publishing their results. They make unrelaistic assumptions, they use questionable approaches to making claims, they don't use enough variables (in the case of this report, they didn't even factor in Nader!) And when they find something they believe is significant, they bypass the scientific method completely, and rush straight to internet blogs or PRNewswire.
Again, let the scientific method take its course, and be very cautious of anything that doesn't.