I'd say blame the media. They're the ones who made more of Gore's "wooden appearance" than of Bush's cocaine use, his many business failures or the fact he dodged Vietnam.
Which media? Surely you don't mean the liberal MSM, do you?;)
Bigelow Aerospace was founded by Robert Bigelow and is funded by the fortune Bigelow gained through his ownership of the hotel chain Budget Suites of America."
Hence the whole idea of a "space hotel" seems appropriate...
Btw, now that one can RTFA (no fault of yours if you didn't read it before), it turns out that there are 4 Republican campaigns running on Linux and 9 running on Windows. As for the Democrats, there are 7* runing on Linux, 1 running on Windows, and 2 running on BSD. That makes this even less likely to be due soley to random chance. (As you'll see elsewhere, under the old assumptions it was already slightly less than 1% due to pure chance.)
This is not to imply some big Windows/Linux controversy. All I am arguing is that it is not purely random. There are lots of possibily non-sinister, non-controversial explanations for how this could come to be, many of which have been discussed elsewhere.
*This includes Al Gore, for some odd reason. In fact, looking at the entire list, there are probably other dubious inclusion. If you want to argue that this is non-meaningful, your best recourse is to argue for cherry-picking the data. As I have no way of knowing how the data were picked, I can't argue effectively against that.:)
The.9% value you arrived at agrees very closely with the 1/104 value that I arrived at by applying Fisher's Exact Test. Since I also derived a totally different value (14%) using a self-invented naive method, I'm inclined to believe that my math in doing the Fisher's Exact Test was not wrong, but that my self-invented naive method was either too naive or suffered from a mathematical mistake itself.
All I was saying was that it seemed unlikely that it was random. Once you've ruled out random, you're free to explore other hypotheses. You'll note there are several already floating out there, but I'll choose to remain agnostic as to what the "real" reason(s) are.
You'll notice that the assumption I preferred was that 1/3 of the hosts ran a MS OS, so that's not far off from 35%.
Also the decision is usually not random but based on how the candidate can get the best deal. In most cases the only reason you would be using microsoft as the host on your website is if your website was being provided to you by a large corporation that has its own servers and uses microsoft; because the management is only familiar with microsoft project and wants to keep everything consistant.
I was arguing exactly that it wasn't random, although I was remaining agnostic as to the rationale.
That's what I was thinking of, although I think it was just called a "Fisher's test" back when I (sort of) learned it. Interestingly enough, when I assume there are 9 Democrats (as I did in my other calculation), I get p=1/104, or about 1%. (For the first "box" I get 9/1768, and for the second "box" (there are only two), I get 1/221, and 9/1768+1/221=1/104. So, either my original assumptions (which also requires an a priori assumption that this does not) were way off, or my calculations here are messed up.
First, let's assume (for sake of argument) that this wasn't a post-hoc analysis (requiring more rigorous analysis). I can't imagine you'd find any journal that would publish results that have a 1-in-7 chance of being the result of "good luck". 1-in-7 is "interesting", as in "deserves further research". It's not significant, at least not to me. (You're fundamentally, right, of course, that what is "significant" is not a hard line that one can draw in the sand.)
And what if I think that 20% or 50% is significant?
Obviously, with a post-hoc analysis, 50% is not just insignificant, it's not even interesting. (It's like flipping a coin, and after the coin lands heads, saying "wow, that coin seems to be biased towards heads".) Even discarding the post-hoc component, you've gone from getting lucky with a die coming up a "6" to a coin coming up "heads". If you do two such studies, you'd expect one of them to be 50% "significant".
Not everyone runs their experiments at the book-generated, easily available 10%, 5%, 1%, etc. levels of statistical signficance. Maybe 15% or 20% will be ok for me.
You may want to look into Nonparametric statistics. Not everyone can take 30 samples and look at the chart on the inner cover of the STAT101 book for the results... (example : how much stress can a 10' gear costing $10,000 take before breaking? I doubt that they will give you 30 of them to break...)
I could definitely do with a statistics refresher (and expander) course. I took a "graduate-level" introduction to statistics course here at UVA (designed for engineers, no less), and it was a joke. The amount of time spent discussing how to calculate the median made me want to gouge my eyes out with my fingers. That said, you'll notice that I did "roll my own distribution" and found it wanting.
What's the name of the proper test for 2-group category data? I tried sifting through Wikipedia to find it, but it's been more than 15 years since I had a (proper) statistics class.
Referring to my other calculations, one could argue that the chances of this happening by chance are about 1 in 7. So, yes, as you say, not statistically significant. What is interesting is that Hillary (arguably the most conservative Democrat) is the one Democratic Windows user, and two of the four Republican Linux users are Ron Paul and John McCain. So, I do not believe it is "pure randomness", but that hypothesis cannot be ruled out. (I.e., you're absolutely correct in every sense.)
I forgot the one I'm thinking of, however. Here's the gist: for sake of argument, let's assume that there are 9 Democratic candidates. If we now naively assume that the odds are 50/50 then the probability of having less than 2 Democrats using Windows is 9/2^9, or approximately 1.8%. If instead, we assume that the odds are 1/3 (the average over both groups), then the probability of less than 2 Democrats using Windows is q^9 + 8pq^8 (p=1/3, q=2/3), or about 13%. So, it doesn't pass the 5% threshold, but it does seem quite unlikely.
You're only commenting on the Republican side. I was referring to the discrepancy between the Democrats (1 out of X) and the Republicans (5 out of 9). As I implied elsewhere, I'm sure a Student's t-test would show this to be quite statistically significant. (Unfortunately, I can't RTFA to figure out what "X" is, but I'm guessing it's also about 9.) Also, your implied assumption is a priori 50/50 odds, which seems like a hell of an assumption. (Comparing the two groups requires no such assumption.)
Interesting that in the short list you provided, you omitted the party that some people credit for causing Gore to lose the 2000 election. (I'm not making that claim, I'm just pointing out the claim.)
I don't plan on doing a students t-test or anything, but these results seem pretty significant to me. You're right that the candidates probably don't know what operating system their web server is running, so instead it speaks more to the kind of people they hire to run their web servers. One can easily make generalizations about both groups, so I'll leave that as an exercise for the reader.
The first place to go to get ideas about applications that can be sufficiently parallelized is to consider all of the @Home programs. In addition to those, there are the ones already mentioned by those who have responded to you. Then there's my research area - neural network models of mammalian brains. One of the things that people plan on doing on Blue Gene is simulating an entire human brain.
I think it would be easier (relatively speaking) to find comets that provided the appropriate concentrations of CO2/H20/etc. to replenish the "leaking" atmosphere than to try to provide enough additional mass and/or magnetic field to retain the atmosphere indefinitely. (OTOH, increasing the gravity serves two purposes of course.)
Her ideas are complex, and have many interesting ramifications - just like Karl Marx. Now I realize that in most ways she was the antithesis of Karl Marx, but IMNSHO she makes the same fundamental mistake as Karl Marx - she appears to place too much faith in her fellow human being. Of course, I've only read one book (Anthem) by her, so I'm definitely no expert on her thinking. I'm basing this mainly on that book (recently read), but I'm also basing it on how her "supporters" describe her philosophy (which largely agrees with what I got out of the book).
Venus, like Mars, also has a very weak magnetic field. It's a combination of factors (including gravity), of which the magnetic field is an important component. Also, the magnetic field protects from long-term stripping of the atmosphere (i.e., over hundreds of millions of years), so as long as we could continue to replenish the atmosphere (at a very low rate), it could work. I'm no planetary specialist, though, so take that with a grain of salt.
Slashdot Mirror
Since the "party" lines would dictate 7-2 splits instead of merely 5-4 splits.
I missed the reference to any law. Which article was that in?
After all, your statment was in this case, indeed, obligatory. (If you hadn't already said it, I was going to.)
Btw, now that one can RTFA (no fault of yours if you didn't read it before), it turns out that there are 4 Republican campaigns running on Linux and 9 running on Windows. As for the Democrats, there are 7* runing on Linux, 1 running on Windows, and 2 running on BSD. That makes this even less likely to be due soley to random chance. (As you'll see elsewhere, under the old assumptions it was already slightly less than 1% due to pure chance.)
This is not to imply some big Windows/Linux controversy. All I am arguing is that it is not purely random. There are lots of possibily non-sinister, non-controversial explanations for how this could come to be, many of which have been discussed elsewhere.
*This includes Al Gore, for some odd reason. In fact, looking at the entire list, there are probably other dubious inclusion. If you want to argue that this is non-meaningful, your best recourse is to argue for cherry-picking the data. As I have no way of knowing how the data were picked, I can't argue effectively against that. :)
The .9% value you arrived at agrees very closely with the 1/104 value that I arrived at by applying Fisher's Exact Test. Since I also derived a totally different value (14%) using a self-invented naive method, I'm inclined to believe that my math in doing the Fisher's Exact Test was not wrong, but that my self-invented naive method was either too naive or suffered from a mathematical mistake itself.
All I was saying was that it seemed unlikely that it was random. Once you've ruled out random, you're free to explore other hypotheses. You'll note there are several already floating out there, but I'll choose to remain agnostic as to what the "real" reason(s) are.
That's what I was thinking of, although I think it was just called a "Fisher's test" back when I (sort of) learned it. Interestingly enough, when I assume there are 9 Democrats (as I did in my other calculation), I get p=1/104, or about 1%. (For the first "box" I get 9/1768, and for the second "box" (there are only two), I get 1/221, and 9/1768+1/221=1/104. So, either my original assumptions (which also requires an a priori assumption that this does not) were way off, or my calculations here are messed up.
I could definitely do with a statistics refresher (and expander) course. I took a "graduate-level" introduction to statistics course here at UVA (designed for engineers, no less), and it was a joke. The amount of time spent discussing how to calculate the median made me want to gouge my eyes out with my fingers. That said, you'll notice that I did "roll my own distribution" and found it wanting.
What's the name of the proper test for 2-group category data? I tried sifting through Wikipedia to find it, but it's been more than 15 years since I had a (proper) statistics class.
Referring to my other calculations, one could argue that the chances of this happening by chance are about 1 in 7. So, yes, as you say, not statistically significant. What is interesting is that Hillary (arguably the most conservative Democrat) is the one Democratic Windows user, and two of the four Republican Linux users are Ron Paul and John McCain. So, I do not believe it is "pure randomness", but that hypothesis cannot be ruled out. (I.e., you're absolutely correct in every sense.)
That last equation should read q^9 + 9pq^8, or about 14%.
I forgot the one I'm thinking of, however. Here's the gist: for sake of argument, let's assume that there are 9 Democratic candidates. If we now naively assume that the odds are 50/50 then the probability of having less than 2 Democrats using Windows is 9/2^9, or approximately 1.8%. If instead, we assume that the odds are 1/3 (the average over both groups), then the probability of less than 2 Democrats using Windows is q^9 + 8pq^8 (p=1/3, q=2/3), or about 13%. So, it doesn't pass the 5% threshold, but it does seem quite unlikely.
You're only commenting on the Republican side. I was referring to the discrepancy between the Democrats (1 out of X) and the Republicans (5 out of 9). As I implied elsewhere, I'm sure a Student's t-test would show this to be quite statistically significant. (Unfortunately, I can't RTFA to figure out what "X" is, but I'm guessing it's also about 9.) Also, your implied assumption is a priori 50/50 odds, which seems like a hell of an assumption. (Comparing the two groups requires no such assumption.)
Pure randomness? I'm sure someone here will be willing to calculate the odds against that for you, if you like...
Interesting that in the short list you provided, you omitted the party that some people credit for causing Gore to lose the 2000 election. (I'm not making that claim, I'm just pointing out the claim.)
I don't plan on doing a students t-test or anything, but these results seem pretty significant to me. You're right that the candidates probably don't know what operating system their web server is running, so instead it speaks more to the kind of people they hire to run their web servers. One can easily make generalizations about both groups, so I'll leave that as an exercise for the reader.
No. However, it was a poorly surveyed area, so non-existence of (prior) evidence is not evidence of (prior) non-existence.
The first place to go to get ideas about applications that can be sufficiently parallelized is to consider all of the @Home programs. In addition to those, there are the ones already mentioned by those who have responded to you. Then there's my research area - neural network models of mammalian brains. One of the things that people plan on doing on Blue Gene is simulating an entire human brain.
I think it would be easier (relatively speaking) to find comets that provided the appropriate concentrations of CO2/H20/etc. to replenish the "leaking" atmosphere than to try to provide enough additional mass and/or magnetic field to retain the atmosphere indefinitely. (OTOH, increasing the gravity serves two purposes of course.)
Her ideas are complex, and have many interesting ramifications - just like Karl Marx. Now I realize that in most ways she was the antithesis of Karl Marx, but IMNSHO she makes the same fundamental mistake as Karl Marx - she appears to place too much faith in her fellow human being. Of course, I've only read one book (Anthem) by her, so I'm definitely no expert on her thinking. I'm basing this mainly on that book (recently read), but I'm also basing it on how her "supporters" describe her philosophy (which largely agrees with what I got out of the book).
When calling other people stupid, it's best to use big words like "too" and "she" correctly.
Venus, like Mars, also has a very weak magnetic field. It's a combination of factors (including gravity), of which the magnetic field is an important component. Also, the magnetic field protects from long-term stripping of the atmosphere (i.e., over hundreds of millions of years), so as long as we could continue to replenish the atmosphere (at a very low rate), it could work. I'm no planetary specialist, though, so take that with a grain of salt.