A side point of the OVC model is that you don't have to verify the code, because you can verify the output instead, since the voting machines themselves don't have any memory of the ballots printed and only the output matters.
Now, I'd be happier if the ballots printed out used machine-readable fonts rather than human-unreadable barcodes, but subverting the OVC model has difficulty on-par with subverting a paper ballot because either (a) you subvert all the barcode readers at a counting location, or (b) you only subvert some of them, in which case counters who get suspicious can cross-check some of the ballots with another barcode reader. Plus, the promoted OVC scenario has a handful of barcode readers available for the voters to cross-check their ballots with; while only a tiny fraction actually will, that tiny fraction should be enough to spot fraud from the manufacturers of the barcode readers.
But at this point, isn't the net effect that your browser-side SSL certificate is really "you"? That means your browser has become you, and when you walk away from your computer, someone else can walk up to it and become you. So you lock your keyboard/account, and then it's back to a password.
I guess at least the remote attack is gone, and only the local attack remains. Unless of course there's some new attack that wrests the certificate from the browser. Then "you" have been duplicated away from your computer.
I don't know about IE or Opera, but if you set a master password in Firefox, then you can't use your SSL certificates until you enter that password. At that point, it's effectively 1.5-factor authentication.
Sure, someone could trojan your box, wait for you to unlock your SSL certificate, then spirit it away; but a trojan in your box could also perform man-in-the-middle attacks on your connection, even if you used phone callbacks, RSA keyfobs, or any other true 2-factor authentication. If you can't trust your local box, then you can't trust your local box, and no amount of magic pixie dust will make a trojaned local box safe to use.
The Egyptians had a far more convincing arguments that not only allowed them to measure the circumference of the Earth's sphere to an amazing accuracy but the distance to the Moon. They achieved this amazing feat about 4000 years ago by measuring the angle of the sun to the Earth at different latitudes simultaneously on the same longitude a known distance apart!
The famous measurement of the Earth's circumference wasn't the doing of the ancient pyramid-building Egyptians that we know and love, but rather an accomplishment of a Greek scholar named Eratosthenes living in Egypt during the Ptolemaic period. This was roughly in 240BC, which was "merely" 2250 years ago, not 4000.
Don't get me wrong: the Egyptians of 4000 years ago had made some fairly solid advances in math -- for instance, they could do basic algebraic proofs and had a workable system for rational numbers and fractions -- but they didn't know the Pythagorean_theorem and weren't very advanced in geometry, trigonometry, or dealing with irrational numbers. This lack of mathematical knowledge would've greatly hindered any attempt they might have made to compute the circumference of the Earth, if they ever actually tried it.
Also, many OpenID providers like MyOpenID let you generate a browser-side SSL certificate and forbid password logins entirely on your account. At that point, you can't be tricked into entering your password because you simply don't have a password.
Santa Claus? What's he going to do if he damns you? Bring you a lump of coal? With the prices of energy lately, being damned by Santa might actually be a blessing these days, if you know what I mean.
Actually, besides being a count of pages matched and not domains (like another reply mentioned), Google's result counts are just estimates and can sometimes be wrong by several orders of magnitude. The numbers get more accurate the deeper you go into the search results, i.e. the first page statement "Results 1-10 of about X" has the most inaccurate value of X that Google returns.
Things like GoogleFight are somewhat useful, in a fun way, since a difference in result counts between two terms probably reflects an actual difference in popularity (albeit completely ignoring PageRank and other spam-unfriendly metrics). However, I cringe every time I see someone cite Google result counts in a serious article or (ugh) research paper, because that's missing the point entirely of what the word "about" is implying.
He claims to have had a radical 180 change of heart since getting the boot from Congress in 2003, including a denouncement of the PATRIOT act, an about-face on medical marijuana, and an apology for the Defense of Marriage Act.
The otherwise farfetched-but-partially-plausible article that Slashdot ran last week had one thing very right: there's no logical reason to believe, based on YHOO's performance over the year prior to Microsoft's first offer, that $19 was anything but a temporary low blip before Microsoft swooped in like a vulture at just the right moment to make their takeover bid look more impressive.
The average YHOO stock price over the preceding year was (eyeballing here) roughly $27 per share, with a general slight downward trend but still high enough that a linear fit would still predict a price around $23-$25 (again, eyeballing) at the time of Microsoft's first public offer. When you compare YHOO to NASDAQ over that same year, it becomes obvious that the timing of YHOO's ups and downs had much more to do with volatility and emotions on NASDAQ (and the larger oil-credit-bear global stock market) than it did with news regarding Yahoo itself. The buys and sells, modulo the bump-and-slump after the layoff announcement on Jan 21, clearly aren't due to new information about Yahoo's fundamentals as a company, so it seems fairly reasonable that the $19 share price (which lasted for a mere 2 days) wouldn't have lasted any longer than the wait for the next ephemeral upward bump in the NASDAQ. (Not that YHOO would've outperformed NASDAQ, necessarily, but it quite likely would've be back to the $21-$23 range soon enough, and possibly higher.)
When you combine this information, it makes Microsoft's $31 deal look much more like a lowball number that it does at first glance, and makes it quite clear that Yahoo's board was reasonable to perceive it as such, even if it turns out they were wrong in the final analysis.
In addition, for most of the last 5 years, YHOO has traded $25 or higher, and sometimes as high as $40. Traders who bought YHOO as a long-term tech investment when it was $25 or higher -- likely the majority of YHOO shareholders -- would've been treated to a much less impressive return in the MS deal, or even a loss depending on the original buy price. For them, the MS deal could easily be beaten by simple share appreciation over 5 to 10 years if Yahoo just manages to get its house in order, even if it's merely to be a more solid runner-up behind Google. Yahoo is, after all, #2 overall and #1 in certain markets when it comes to search, and they own Overture, the only company that's been doing online text ad auctions for longer than Google, so they clearly have the potential for a turnaround. This whole Microsoft fiasco might be the kick in the pants they needed to make it actually happen.
So, in the long analysis for all those investors who bought before YHOO reached $19, it's not even remotely a guarantee that they would've been happy with Microsoft's $31 numbers, or that they're upset the Microsoft deal fell through on the terms that it did. Anyone who says "I poached YHOO at $19 and got ripped off because I didn't get my 63% return" is a moron short-term gambler who deserves to get burned.
What I can't understand for the life of me is why ISPs don't sell QoS as a metered service. "Hello, ma'am, you want this SSH connection to run at AF2? Sure thing, the fee comes to $0.0X per thousand packets. Pleasure doing business. Oh, good day sir, you say you want to run that BitTorrent at EF? Wow, are you sure? The fee is $0.YY per thousand packets, so for 4.5GB and a 1:1 upload ratio, that adds up to $ZZZ.ZZ. Sir? Sir? Are you okay, sir?"
Even if QoS never comes to the Internet backbone at large, just having it within the ISP network would be a vast improvement for things that need better latency guarantees. The downside would be that you can't really trust the QoS bits coming off the end user's computer, since they're set deep in the bowels of applications and not user selectable -- to say nothing of the problem of malware racking up an unexpected bill -- so you'd have to come up with some other mechanism for users to choose and consent.
I call bullshit. Neuron action potentials occur because of ionized atoms (H+, Na+, K+) entering and leaving the cell membrane. There's no such thing as half an atom, or a fractionally charged ion; therefore, the neuron's internal environment is not continuously variable, and thus action potentials are not analog. Quantum mechanics has shown that infinitely graduated anything is impossible in the physical world.
Are action potentials more subtle than a binary toggle? Sure. But that doesn't make them analog.
Thank you. I had thought most of the alcohol effect was osmotic since some yeasts thrive up to about 18%v. Does alcohol work by esterification displacement of the amino bonds?
There's probably some of that, but at room temperature probably not enough to matter. Most of it is just because the tertiary shape of a protein is almost purely determined by hydrogen-bonding interactions. When you introduce a solvent less polar than water that is nonetheless miscible with it, the result is that the hydrophobic "core" of the protein can start to unravel. All the amino acids are still in the right order, but the protein is useless because it's no longer folded into the correct overall shape.
... But if you switch the device on long enough to measure 1000 photons, the pattern, or lack thereof, would be clear as day...
What pattern?
Say you, Alice, and Bob are measuring polarization of photons along an axis in an attempt to create FTL communication. You have a source that emits pairs of photons, quantum entangled such that the two photons in each pair have unknown but exactly opposite polarizations, which is a typical setup for a QM experiment. You separate the outgoing stream of photon pairs, sending one from each pair to Alice and another to Bob. Alice has a detector that measures each photon into "polarization matches this axis" (Y) and "polarization does not match this axis" (N). Bob has a second detector that can perform the same measurement, on the same axis. Alice and Bob are far enough away that, although Alice measures her photon first, Alice doesn't have time to send a message about her result to Bob before Bob measures his photon. (We can say "first" and "before" with confidence, even though this is a Special Relativistic context involving information travel at light speed, because Alice and Bob are at rest with respect to each other.)
So you send out 1000 pairs of photons. Alice sees a stream of photons (N Y Y N Y N N Y Y N Y...), and then Bob sees a stream of photons (Y N N Y N Y Y N N Y N...). Now Alice and Bob switch, and Bob reads first. You send another 1000 pairs of photons. Bob sees a stream of photons (Y N Y Y N N N Y N Y Y...), and then Alice sees a stream of photons (N Y N N Y Y Y N Y N N...).
Now what? When Alice measures hers first, she knows what Bob will see when he measures his, and vice versa when they swap. But, other than each stream being roughly 50% Y and 50% N, there's no pattern to be seen.
This experiment setup doesn't even bother to demonstrate QM's "spooky action at a distance"; it's true even in classical physics. But it thoroughly demonstrates why there's no FTL communication possible: when Alice measured first, she had no power to choose what her result (and thus Bob's) would be, and when she and Bob switched, there was no difference whatsoever.
This continues to be true when the experiment is modified to invoke QM's "spooky action", by having Alice and Bob randomly pick between two polarization axes, because when Bob looks at his data in isolation, there's still no pattern until he sees Alice's as well. And when they compare finally notes, they will notice that whenever they picked the same polarization axis, they always got opposite results, but when they picked different polarization axes, they got completely unrelated results. No FTL communication.
Actually, there's pretty good evidence now that HPV is also responsible for certain mouth, throat, penis, and anus cancers. The sheer number of mouth and throat cancers caused by tobacco (chewing and smoking are both dangerous) puts HPV to shame, but unprotected oral sex (with either gender) can spread HPV to the mouth and throat.
Also, HPV isn't responsible for infertility, except in the sense that it can cause cancers that require fertility-damaging surgery to remove. Most of the STDs that cause infertility are bacterial, not viral, and non-cancerous HPV infections are almost entirely harmless.
Actually, it does, it's just most people die of AIDS before the HIV infection itself can kill. In addition to infecting immune system cells, HIV also infects brain cells, which is the cause of the cognitive impairment descending into dementia that is experienced by a significant number of late-stage AIDS patients. If the patient doesn't die of infection first, AIDS dementia complex can destroy enough brain tissue to finish them off.
This is a completely outrageous lie. Copyright doesn't automatically transfer just because you click "upload", and if any company tried burying that in the Terms of Service, a judge would laugh them out of court with a multi-million-dollar judgment hanging around their neck. You need to sign a contract, in paper and ink like the form letter that the FSF asks people to fill out, sign, and mail to them, and you need to keep that paper contract on file indefinitely in case the previous owner has a change of heart and tries to sue you.
Technically, what you describe isn't analog (i.e. continuous and infinitely differentiable); it's digital, except probabilistic rather than the more familiar deterministic.
I think that, most of the time anyway, "law" as used in science has an even more specific meaning: a "law" is a relation (often an equation) between two or more variables. For instance, Boyle's Law states "for a fixed amount of gas kept at a fixed temperature, pressure and volume are inversely proportional". This is, strictly speaking, not true of reality. It describes an ideal gas with completely elastic collisions, a property that no real gas has. But it's close enough to true with real gases that it offers a good guess of how a real gas will behave.
Newton's Laws are used similarly. No reasonable person still accepts Newton's Theories of Motion and Gravity, because Einstein's two Theories of Relativity have supplanted them and have thoroughly demonstrated their predictive power. However, Newton's Laws of Motion and Gravity are still taught to students. Laws are not "correct" vs. "incorrect", because they're abstract mathematical relationships; instead, laws get sorted into "useful" vs. "not useful" categories, and Newton's Laws are good enough at estimating reality that they're still useful. But they won't stop being "correct" any more than "f(x)=x^2" will stop being "correct", because a law continues to be a law even when no surviving theory references it.
Actually, alcohol is a wonderful agent for denaturing proteins, and if you disrupt the viral capsid by denaturing it, it's no longer capable of attaching to host cells and causing infection.
Approval and Range (the former being a special case of the latter) are even better, and both sidestep Arrow's Impossibility Theorem by not being ranked ballots.
As a fairly new Google employee, who is now bound by NDAs and thus probably can't say anything about our development process, I have chosen an alternate means of expression:
... and I doubt that many write unit tests.
*grumble*
*grumble* *grumble*
*grumble* *grumble* *grumble*
This concludes the unit test of the Emergency Grumbling System.
Slashdot Mirror
Does the "--no-remote" command line flag not work for you?
A side point of the OVC model is that you don't have to verify the code, because you can verify the output instead, since the voting machines themselves don't have any memory of the ballots printed and only the output matters.
Now, I'd be happier if the ballots printed out used machine-readable fonts rather than human-unreadable barcodes, but subverting the OVC model has difficulty on-par with subverting a paper ballot because either (a) you subvert all the barcode readers at a counting location, or (b) you only subvert some of them, in which case counters who get suspicious can cross-check some of the ballots with another barcode reader. Plus, the promoted OVC scenario has a handful of barcode readers available for the voters to cross-check their ballots with; while only a tiny fraction actually will, that tiny fraction should be enough to spot fraud from the manufacturers of the barcode readers.
Whoops, I'll have to remember that.
As a longtime B5 viewer, I can confirm that the poster making the "Picabo" claim is pulling your leg.
I don't know about IE or Opera, but if you set a master password in Firefox, then you can't use your SSL certificates until you enter that password. At that point, it's effectively 1.5-factor authentication.
Sure, someone could trojan your box, wait for you to unlock your SSL certificate, then spirit it away; but a trojan in your box could also perform man-in-the-middle attacks on your connection, even if you used phone callbacks, RSA keyfobs, or any other true 2-factor authentication. If you can't trust your local box, then you can't trust your local box, and no amount of magic pixie dust will make a trojaned local box safe to use.
The famous measurement of the Earth's circumference wasn't the doing of the ancient pyramid-building Egyptians that we know and love, but rather an accomplishment of a Greek scholar named Eratosthenes living in Egypt during the Ptolemaic period. This was roughly in 240BC, which was "merely" 2250 years ago, not 4000.
Don't get me wrong: the Egyptians of 4000 years ago had made some fairly solid advances in math -- for instance, they could do basic algebraic proofs and had a workable system for rational numbers and fractions -- but they didn't know the Pythagorean_theorem and weren't very advanced in geometry, trigonometry, or dealing with irrational numbers. This lack of mathematical knowledge would've greatly hindered any attempt they might have made to compute the circumference of the Earth, if they ever actually tried it.
Also, many OpenID providers like MyOpenID let you generate a browser-side SSL certificate and forbid password logins entirely on your account. At that point, you can't be tricked into entering your password because you simply don't have a password.
Hey, man, don't mess with the Krampus!
I start getting that one at around the 24 hour mark. It weirds me out enough that I've never gone more than about 28 or 30 hours without sleep.
Actually, besides being a count of pages matched and not domains (like another reply mentioned), Google's result counts are just estimates and can sometimes be wrong by several orders of magnitude. The numbers get more accurate the deeper you go into the search results, i.e. the first page statement "Results 1-10 of about X" has the most inaccurate value of X that Google returns.
Things like GoogleFight are somewhat useful, in a fun way, since a difference in result counts between two terms probably reflects an actual difference in popularity (albeit completely ignoring PageRank and other spam-unfriendly metrics). However, I cringe every time I see someone cite Google result counts in a serious article or (ugh) research paper, because that's missing the point entirely of what the word "about" is implying.
He claims to have had a radical 180 change of heart since getting the boot from Congress in 2003, including a denouncement of the PATRIOT act, an about-face on medical marijuana, and an apology for the Defense of Marriage Act.
Not sure if I believe him, but...
The otherwise farfetched-but-partially-plausible article that Slashdot ran last week had one thing very right: there's no logical reason to believe, based on YHOO's performance over the year prior to Microsoft's first offer, that $19 was anything but a temporary low blip before Microsoft swooped in like a vulture at just the right moment to make their takeover bid look more impressive.
The average YHOO stock price over the preceding year was (eyeballing here) roughly $27 per share, with a general slight downward trend but still high enough that a linear fit would still predict a price around $23-$25 (again, eyeballing) at the time of Microsoft's first public offer. When you compare YHOO to NASDAQ over that same year, it becomes obvious that the timing of YHOO's ups and downs had much more to do with volatility and emotions on NASDAQ (and the larger oil-credit-bear global stock market) than it did with news regarding Yahoo itself. The buys and sells, modulo the bump-and-slump after the layoff announcement on Jan 21, clearly aren't due to new information about Yahoo's fundamentals as a company, so it seems fairly reasonable that the $19 share price (which lasted for a mere 2 days) wouldn't have lasted any longer than the wait for the next ephemeral upward bump in the NASDAQ. (Not that YHOO would've outperformed NASDAQ, necessarily, but it quite likely would've be back to the $21-$23 range soon enough, and possibly higher.)
When you combine this information, it makes Microsoft's $31 deal look much more like a lowball number that it does at first glance, and makes it quite clear that Yahoo's board was reasonable to perceive it as such, even if it turns out they were wrong in the final analysis.
In addition, for most of the last 5 years, YHOO has traded $25 or higher, and sometimes as high as $40. Traders who bought YHOO as a long-term tech investment when it was $25 or higher -- likely the majority of YHOO shareholders -- would've been treated to a much less impressive return in the MS deal, or even a loss depending on the original buy price. For them, the MS deal could easily be beaten by simple share appreciation over 5 to 10 years if Yahoo just manages to get its house in order, even if it's merely to be a more solid runner-up behind Google. Yahoo is, after all, #2 overall and #1 in certain markets when it comes to search, and they own Overture, the only company that's been doing online text ad auctions for longer than Google, so they clearly have the potential for a turnaround. This whole Microsoft fiasco might be the kick in the pants they needed to make it actually happen.
So, in the long analysis for all those investors who bought before YHOO reached $19, it's not even remotely a guarantee that they would've been happy with Microsoft's $31 numbers, or that they're upset the Microsoft deal fell through on the terms that it did. Anyone who says "I poached YHOO at $19 and got ripped off because I didn't get my 63% return" is a moron short-term gambler who deserves to get burned.
What I can't understand for the life of me is why ISPs don't sell QoS as a metered service. "Hello, ma'am, you want this SSH connection to run at AF2? Sure thing, the fee comes to $0.0X per thousand packets. Pleasure doing business. Oh, good day sir, you say you want to run that BitTorrent at EF? Wow, are you sure? The fee is $0.YY per thousand packets, so for 4.5GB and a 1:1 upload ratio, that adds up to $ZZZ.ZZ. Sir? Sir? Are you okay, sir?"
Even if QoS never comes to the Internet backbone at large, just having it within the ISP network would be a vast improvement for things that need better latency guarantees. The downside would be that you can't really trust the QoS bits coming off the end user's computer, since they're set deep in the bowels of applications and not user selectable -- to say nothing of the problem of malware racking up an unexpected bill -- so you'd have to come up with some other mechanism for users to choose and consent.
I call bullshit. Neuron action potentials occur because of ionized atoms (H+, Na+, K+) entering and leaving the cell membrane. There's no such thing as half an atom, or a fractionally charged ion; therefore, the neuron's internal environment is not continuously variable, and thus action potentials are not analog. Quantum mechanics has shown that infinitely graduated anything is impossible in the physical world.
Are action potentials more subtle than a binary toggle? Sure. But that doesn't make them analog.
There's probably some of that, but at room temperature probably not enough to matter. Most of it is just because the tertiary shape of a protein is almost purely determined by hydrogen-bonding interactions. When you introduce a solvent less polar than water that is nonetheless miscible with it, the result is that the hydrophobic "core" of the protein can start to unravel. All the amino acids are still in the right order, but the protein is useless because it's no longer folded into the correct overall shape.
Do not taint me by comparing me to Carl Icahn.
Prick.
What pattern?
Say you, Alice, and Bob are measuring polarization of photons along an axis in an attempt to create FTL communication. You have a source that emits pairs of photons, quantum entangled such that the two photons in each pair have unknown but exactly opposite polarizations, which is a typical setup for a QM experiment. You separate the outgoing stream of photon pairs, sending one from each pair to Alice and another to Bob. Alice has a detector that measures each photon into "polarization matches this axis" (Y) and "polarization does not match this axis" (N). Bob has a second detector that can perform the same measurement, on the same axis. Alice and Bob are far enough away that, although Alice measures her photon first, Alice doesn't have time to send a message about her result to Bob before Bob measures his photon. (We can say "first" and "before" with confidence, even though this is a Special Relativistic context involving information travel at light speed, because Alice and Bob are at rest with respect to each other.)
So you send out 1000 pairs of photons. Alice sees a stream of photons (N Y Y N Y N N Y Y N Y ...), and then Bob sees a stream of photons (Y N N Y N Y Y N N Y N ...). Now Alice and Bob switch, and Bob reads first. You send another 1000 pairs of photons. Bob sees a stream of photons (Y N Y Y N N N Y N Y Y ...), and then Alice sees a stream of photons (N Y N N Y Y Y N Y N N ...).
Now what? When Alice measures hers first, she knows what Bob will see when he measures his, and vice versa when they swap. But, other than each stream being roughly 50% Y and 50% N, there's no pattern to be seen.
This experiment setup doesn't even bother to demonstrate QM's "spooky action at a distance"; it's true even in classical physics. But it thoroughly demonstrates why there's no FTL communication possible: when Alice measured first, she had no power to choose what her result (and thus Bob's) would be, and when she and Bob switched, there was no difference whatsoever.
This continues to be true when the experiment is modified to invoke QM's "spooky action", by having Alice and Bob randomly pick between two polarization axes, because when Bob looks at his data in isolation, there's still no pattern until he sees Alice's as well. And when they compare finally notes, they will notice that whenever they picked the same polarization axis, they always got opposite results, but when they picked different polarization axes, they got completely unrelated results. No FTL communication.
Actually, there's pretty good evidence now that HPV is also responsible for certain mouth, throat, penis, and anus cancers. The sheer number of mouth and throat cancers caused by tobacco (chewing and smoking are both dangerous) puts HPV to shame, but unprotected oral sex (with either gender) can spread HPV to the mouth and throat.
Also, HPV isn't responsible for infertility, except in the sense that it can cause cancers that require fertility-damaging surgery to remove. Most of the STDs that cause infertility are bacterial, not viral, and non-cancerous HPV infections are almost entirely harmless.
Actually, it does, it's just most people die of AIDS before the HIV infection itself can kill. In addition to infecting immune system cells, HIV also infects brain cells, which is the cause of the cognitive impairment descending into dementia that is experienced by a significant number of late-stage AIDS patients. If the patient doesn't die of infection first, AIDS dementia complex can destroy enough brain tissue to finish them off.
This is a completely outrageous lie. Copyright doesn't automatically transfer just because you click "upload", and if any company tried burying that in the Terms of Service, a judge would laugh them out of court with a multi-million-dollar judgment hanging around their neck. You need to sign a contract, in paper and ink like the form letter that the FSF asks people to fill out, sign, and mail to them, and you need to keep that paper contract on file indefinitely in case the previous owner has a change of heart and tries to sue you.
Technically, what you describe isn't analog (i.e. continuous and infinitely differentiable); it's digital, except probabilistic rather than the more familiar deterministic.
I think that, most of the time anyway, "law" as used in science has an even more specific meaning: a "law" is a relation (often an equation) between two or more variables. For instance, Boyle's Law states "for a fixed amount of gas kept at a fixed temperature, pressure and volume are inversely proportional". This is, strictly speaking, not true of reality. It describes an ideal gas with completely elastic collisions, a property that no real gas has. But it's close enough to true with real gases that it offers a good guess of how a real gas will behave.
Newton's Laws are used similarly. No reasonable person still accepts Newton's Theories of Motion and Gravity, because Einstein's two Theories of Relativity have supplanted them and have thoroughly demonstrated their predictive power. However, Newton's Laws of Motion and Gravity are still taught to students. Laws are not "correct" vs. "incorrect", because they're abstract mathematical relationships; instead, laws get sorted into "useful" vs. "not useful" categories, and Newton's Laws are good enough at estimating reality that they're still useful. But they won't stop being "correct" any more than "f(x)=x^2" will stop being "correct", because a law continues to be a law even when no surviving theory references it.
Actually, alcohol is a wonderful agent for denaturing proteins, and if you disrupt the viral capsid by denaturing it, it's no longer capable of attaching to host cells and causing infection.
Ew, no. IRV is needlessly complex, too easy to game, and swings around unpredictably between "2 big parties 1 little party" to "1 big party 2 little parties", resulting in no actual IRV-using countries with more than 2 parties/blocs/coalitions. Condorcet is nicer, but suffers from the "Dark Horse" problem.
Approval and Range (the former being a special case of the latter) are even better, and both sidestep Arrow's Impossibility Theorem by not being ranked ballots.
As a fairly new Google employee, who is now bound by NDAs and thus probably can't say anything about our development process, I have chosen an alternate means of expression:
*grumble*
*grumble* *grumble*
*grumble* *grumble* *grumble*
This concludes the unit test of the Emergency Grumbling System.