The difference would probably be that Linux is not uncommon for either research or study, and that
a university that cannot provide support to the point of supporting a sizeable minority of students using
Linux-based systems probably also has inferior programs in areas where Linux would normally be
used (CS, EE, etc.).
But it seems more flawed from the point of view of a group which deliberately makes itself all but impossible to measure.
Given such resources as Debian's popularity contest, it seems at least conceivable that deliberate effort in avoiding measurement may be less attributable to measurement itself than to the particular means of measurement employed.
... then perhaps the best thing to do is to go to the polls.
After all, you probably have a better idea of what your neighbors want than does some fellow across town, so if you believe that your neighbors know better than some fellow across town, then go out and vote with them. Of course, the same argument could be applied to any other group that you are politically aligned with, such as tech-knowledgeable individuals, friends, etc.
If you do decide to stay at home, rest assured that this will probably have no effect on the electoral outcome, but, on the other hand, your friends who do vote might think ill of you if they learn this, since they expended the effort to vote, while you, who are politically aligned with them, didn't, hence reducing their effective voting power.
'Finding the one "left-handed" image from five "right-handed" images below proved difficult, and the Mundurukú study participants did not do much better than chance.'
'Only 23 percent chose the bottom right as the weird or strange image.'
From 6 choices, this is still about 40% better than chance.
Individuals aside, corporations seem quite willing to engage in any and all kinds of inconsistency and unconcern to a degree that would put the most ambitious and selfish individuals to shame. Worse, they're still taken seriously even when caught in the most egregious lies. Would any (non-corporate) person be capable of pulling off SCO's absurd campaign against Linux and IBM?
"Digital cameras are like Macs - they 'just work'."
This is not true for specialized applications. For example, it is generally not straightforward to automatically capture sequences of exposures with varying shutter speeds and aperture sizes. Also, certain information available to the camera, such as characterization of noise, is generally not trivially accessible.
"I haven't heard of any efforts to customize them, or build an open one."
Corey Manders, at the EyeTap lab at the University of Toronto, is developing a custom library for the Nikon D2H digital SLR. Limited details are here:
'Herns told federal agents he was looking for computer space to store movies he'd downloaded. It took hours for technicians to find the problem, fix it and patch the system's security holes, officials said.'
What 'problem' is being referred to in this sentence? Does the reporter not have the backbone to take the most tentative steps toward investigating why unauthorized access had been allowed?
If a physical media copyright infringer had been habitually storing his wares in restricted areas of the Library of Congress, would the AP reporter have written, "It took LoC staff hours to find the problem, fix it and patch the library's security holes"?
More to the point, fixing security problems cannot reasonably be considered damages unless the problems were caused in some way by the unauthorized access (compromised utilities, etc.).
'How did you write a program so fast to find the answre quickly. I don't know much about cryptoanalysis. I found a couple of them, but only by chance...'
Here's what I did for the first part; there may be easier approaches.
First, write a program to transform sequences of characters into a canonical form. Whitespace maps to itself, the first letter in the sequence maps to 'a', the second (unique) letter maps to 'b', etc. For example,
XCT GSXXJYUQ --> abc deaafghi
Then, process GCHQ's lists and IMDB's lists (the latter processed to read 'firstname lastname') into canonical form, and match canonical forms between the two lists. Several IMDB entries may match each GCHQ entry, offering several possible mappings between ciphertext and plaintext. For example,
Map1: XCT GSXXJYUQ --> Rex Harrison
Map2: XCT GSXXJYUQ --> Ted Mattison
Finally, for each ciphertext entry in the list of men, determine which ciphertext->plaintext mappings are consistent with at least one mapping in the list of women. It turns out that this is sufficient to narrow down the number of mappings to one per name, and, as a side effect, it also provides the man/woman pairings. The grandparent post explicitly provides the results, so I won't list examples here.
'At the same time, this isn't a system that can evolve naturally from current systems...'
The proxy/recovery approach you describe could probably be applied in other contexts. (Investment decision-making comes to mind, but it might not be a terribly good example.) Once the technology is proven for other applications, adapting it to function as the decision-making process of a large political system would probably be an easier task.
Continuous elections may... be organized through a system of automatic voting machines similar in principle to an automatic teller machine. Instead of depositing or withdrawing money from a bank account, each voter would be depositing or withdrawing his vote from a particular party or candidate.
I'd probably prefer a condorcet-style ranked election method over the plurality method outlined on the page cited above, however.
... I immediately assumed that the benefits of uncensored information had been recognized as a boon to education. Perhaps I was suffering from momentary naivete, but this interpretation still seems to hold much more promise than the other.
While it may be the case, as you suggest, that the resignations were motivated by personal satisfaction with future prospects for the country, or fear of future prospects, the possibility that they were in fact motivated by the past performance of the administration, or by their own past performance, should perhaps not be too hastily ruled out.
I use light green on dark green. I prefer light-on-dark, and a 100% green color palette should be immune to the near-ubiquitous color convergence problems that CRTs have. Although I rarely use CRTs now, I still use the same colors.
The mechanism by which selection of a candidate for office occurs can be analyzed into parts, and the expected effect of a given vote can be predicted based on how those parts work. Taking this mechanism into account, and voting according to one's best interests is also called 'sophisticated voting'. So, no, I won't say 'you're throwing away your vote', but analyzing the expected effect of a vote is perhaps not an entirely unwarranted exercise.
"If you evaluate your voting power by dividing the number of votes assigned to your state by its population, you're doing it wrong. Voting power indices are not calculated this way. Read about Banzhaf index."
It has been suggested that voting power indices such as the Banzhaf index do not reliably reflect voting power. In particular, the Banzhaf index relies on the random voting model, in which all votes are equally likely, but it is not obvious that this assumption yields correct results. Your larger point is correct, however. Dividing electoral votes by population is not the right way to calculate voting power.
... there must be more interesting ways of evaluating voting methods than determining what effect their adoption would have had on the 2000 election.
E.g., I would be curious to know what effect this system would have on voting power (i.e., the probability that a given vote is decisive). Obviously, the exact voting power would vary by state, but perhaps some interesting general statement could still be made.
"Ok ok, it is overlooked for a very sound reason, continuous voting requires the election to be constantly held, this is difficult in our physical world."
This site suggests ATM-style machines, which Diebold would probably quite happily endorse. The idea had occurred to me before, but I wasn't aware that it had a name. In any case, I think it's a good idea. For elective offices that, for whatever reason, have associated term limits, it should mitigate somewhat the 'lame duck' problem, wherein candidates have no further accountability to voters at all.
(I don't understand why the site linked above emphasizes a combination of continuous voting and plurality voting, however. I would think that Condorcet would be preferable. For stability, there could be a requirement that the incumbent be ejected from the Smith set.)
Another interesting lottery-based approach is Random Ballot. It has the nice property that sincerely voting for the most desirable candidate is also the best strategy.
"Oh, one more caveat. You must realize that *no* alternative voting system can make the US Presidential election fairer for minor parties as long as the Electoral College is in place."
Are there web resources that clearly explain the technical details of how to arrive at this conclusion? While I suspect you are correct, it would be nice to have a URL that I could point people toward.
Also, what are your thoughts on the assertion that combining presidentialism and multipartism leads to political instability? E.g.:
Slashdot Mirror
The difference would probably be that Linux is not uncommon for either research or study, and that a university that cannot provide support to the point of supporting a sizeable minority of students using Linux-based systems probably also has inferior programs in areas where Linux would normally be used (CS, EE, etc.).
... then perhaps the best thing to do is to go to the polls.
After all, you probably have a better idea of what your neighbors want than does some fellow across town, so if you believe that your neighbors know better than some fellow across town, then go out and vote with them. Of course, the same argument could be applied to any other group that you are politically aligned with, such as tech-knowledgeable individuals, friends, etc.
If you do decide to stay at home, rest assured that this will probably have no effect on the electoral outcome, but, on the other hand, your friends who do vote might think ill of you if they learn this, since they expended the effort to vote, while you, who are politically aligned with them, didn't, hence reducing their effective voting power.
More recent PREEMPT_RT benchmarks from the I-PIPE testing team, via Paul McKenney's earlier RT overview.
'Finding the one "left-handed" image from five "right-handed" images below proved difficult, and the Mundurukú study participants did not do much better than chance.'
'Only 23 percent chose the bottom right as the weird or strange image.'
From 6 choices, this is still about 40% better than chance.
Individuals aside, corporations seem quite willing to engage in any and all kinds of inconsistency and unconcern to a degree that would put the most ambitious and selfish individuals to shame. Worse, they're still taken seriously even when caught in the most egregious lies. Would any (non-corporate) person be capable of pulling off SCO's absurd campaign against Linux and IBM?
Colonization plans are proceeding swiftly, I see.
"Digital cameras are like Macs - they 'just work'."
This is not true for specialized applications. For example, it is generally not straightforward to automatically capture sequences of exposures with varying shutter speeds and aperture sizes. Also, certain information available to the camera, such as characterization of noise, is generally not trivially accessible.
"I haven't heard of any efforts to customize them, or build an open one."
Corey Manders, at the EyeTap lab at the University of Toronto, is developing a custom library for the Nikon D2H digital SLR. Limited details are here:
http://www.google.com/search?q=libd2h
'Herns told federal agents he was looking for computer space to store movies he'd downloaded. It took hours for technicians to find the problem, fix it and patch the system's security holes, officials said.'
What 'problem' is being referred to in this sentence? Does the reporter not have the backbone to take the most tentative steps toward investigating why unauthorized access had been allowed?
If a physical media copyright infringer had been habitually storing his wares in restricted areas of the Library of Congress, would the AP reporter have written, "It took LoC staff hours to find the problem, fix it and patch the library's security holes"?
'More like breaking into a bank vault to store the bicycle you just stole.'
... or the set of encyclopedias you just copied.
More to the point, fixing security problems cannot reasonably be considered damages unless the problems were caused in some way by the unauthorized access (compromised utilities, etc.).
... after the accused stole my $3.59 flowerpot, I had to spend hundreds of dollars putting locks on all of my doors.
Here's what I did for the first part; there may be easier approaches.
First, write a program to transform sequences of characters into a canonical form. Whitespace maps to itself, the first letter in the sequence maps to 'a', the second (unique) letter maps to 'b', etc. For example,
XCT GSXXJYUQ --> abc deaafghi
Then, process GCHQ's lists and IMDB's lists (the latter processed to read 'firstname lastname') into canonical form, and match canonical forms between the two lists. Several IMDB entries may match each GCHQ entry, offering several possible mappings between ciphertext and plaintext. For example,
Map1: XCT GSXXJYUQ --> Rex Harrison
Map2: XCT GSXXJYUQ --> Ted Mattison
Finally, for each ciphertext entry in the list of men, determine which ciphertext->plaintext mappings are consistent with at least one mapping in the list of women. It turns out that this is sufficient to narrow down the number of mappings to one per name, and, as a side effect, it also provides the man/woman pairings. The grandparent post explicitly provides the results, so I won't list examples here.
'At the same time, this isn't a system that can evolve naturally from current systems...'
The proxy/recovery approach you describe could probably be applied in other contexts. (Investment decision-making comes to mind, but it might not be a terribly good example.) Once the technology is proven for other applications, adapting it to function as the decision-making process of a large political system would probably be an easier task.
This could be a step towards continuous voting:
I'd probably prefer a condorcet-style ranked election method over the plurality method outlined on the page cited above, however.
... I immediately assumed that the benefits of uncensored information had been recognized as a boon to education. Perhaps I was suffering from momentary naivete, but this interpretation still seems to hold much more promise than the other.
While it may be the case, as you suggest, that the resignations were motivated by personal satisfaction with future prospects for the country, or fear of future prospects, the possibility that they were in fact motivated by the past performance of the administration, or by their own past performance, should perhaps not be too hastily ruled out.
I use light green on dark green. I prefer light-on-dark, and a 100% green color palette should be immune to the near-ubiquitous color convergence problems that CRTs have. Although I rarely use CRTs now, I still use the same colors.
The mechanism by which selection of a candidate for office occurs can be analyzed into parts, and the expected effect of a given vote can be predicted based on how those parts work. Taking this mechanism into account, and voting according to one's best interests is also called 'sophisticated voting'. So, no, I won't say 'you're throwing away your vote', but analyzing the expected effect of a vote is perhaps not an entirely unwarranted exercise.
"If you evaluate your voting power by dividing the number of votes assigned to your state by its population, you're doing it wrong. Voting power indices are not calculated this way. Read about Banzhaf index."
It has been suggested that voting power indices such as the Banzhaf index do not reliably reflect voting power. In particular, the Banzhaf index relies on the random voting model, in which all votes are equally likely, but it is not obvious that this assumption yields correct results. Your larger point is correct, however. Dividing electoral votes by population is not the right way to calculate voting power.
... there must be more interesting ways of evaluating voting methods than determining what effect their adoption would have had on the 2000 election.
E.g., I would be curious to know what effect this system would have on voting power (i.e., the probability that a given vote is decisive). Obviously, the exact voting power would vary by state, but perhaps some interesting general statement could still be made.
"Instant Runoff sounds so nice, so American, so instant!"
Mmm... okay. How about Smith/IRV, an ambiguity resolution technique for Condorcet that limits IRV to the Smith set?
This site suggests ATM-style machines, which Diebold would probably quite happily endorse. The idea had occurred to me before, but I wasn't aware that it had a name. In any case, I think it's a good idea. For elective offices that, for whatever reason, have associated term limits, it should mitigate somewhat the 'lame duck' problem, wherein candidates have no further accountability to voters at all.
(I don't understand why the site linked above emphasizes a combination of continuous voting and plurality voting, however. I would think that Condorcet would be preferable. For stability, there could be a requirement that the incumbent be ejected from the Smith set.)
Another interesting lottery-based approach is Random Ballot. It has the nice property that sincerely voting for the most desirable candidate is also the best strategy.
Are there web resources that clearly explain the technical details of how to arrive at this conclusion? While I suspect you are correct, it would be nice to have a URL that I could point people toward.
Also, what are your thoughts on the assertion that combining presidentialism and multipartism leads to political instability? E.g.:
Party Fragmentation and Presidential Elections in Post-Communist Democracies
(The topic of the paper is somewhat tangential, but it's the best web reference I have found.)