Well, dashing off a memo (one way or the other) doesn't take a lot of time.
What does take a lot of time, cumulatively speaking is reading such memos. If you send a memo and it gets distributed to a couple of hundred people, the cost of that memo can be staggering.
This is a "it depends" question. If you wrap your device in aluminum foil it can become capacitively coupled to the foil. You could fix that by putting it in a room completely lined with foil (increasing the thickness of the dialectric). Or you can connect your foil to a massive electron sink, like a large hunk of metal or the Earth.
Alternatively they could build a sun dial out of masonry, which is a cheap and long-lived material. That would keep time (during the day) as long as the structure existed.
Build it big enough and it could be quite precise, like this one.
As a numbers-oriented guy, it pains me to say this, but Clinton relied on the numbers *too* much.
All polls are adjusted by some kind of likely-voter model; all the "margin of error" figures you hear only take into account *random sampling* errors, not systematic ones. As useful as empirical numbers are, you have to keep in mind that they're only good as the assumptions under which they were collected.
This means that when Clinton got word of something happening that wasn't reflected in her internal polling, she should have taken it seriously rather than discounting it. It was incompetent, but in the opposite way that people are numerically incompetent. People usually trust their gut and discount hard data. But it's possible to trust hard data too much too.
By the way I don't have any problem with Russians or anyone else weighing in on our election -- as long as they do it above board. If they can, as Russians, sway American opinions, more power to them. Pretending to be Americans, to the point of stealing American identities is a different matter.
Again let me reiterate, the goal of Russian efforts seem to have been to undermine Americans' faith in their own country. It's quite possible they had no *intent* of swaying the election, what they may have wanted to do is inflame opposition to a Clinton presidency. Russia has a complicated relationship to the west, because Russia is a country run by economic parasites. Those parasite need stable democratic countries as place to store the wealth they've looted from their own country. Russian oligarchs are behind the insane real estate prices in places like New York, London, Vancouver, and Miami.
Yes, turnout is where the big swings happen in a country where about 40% of the electorate sit out each election; it's not always the same 40%.
Rural turnout was a big factor in 2016, and this speaks to the strengths and weaknesses of the candidates. Clinton, unlike her husband, wasn't good at projecting empathy to voters in rural districts ravaged by economic decline and the opioid crisis. Neither was Trump, but he was much better at projecting their *anger*.
This also shows why Russian meddling may have been quite effective. A lot of it is aimed at spreading disillusionment; it's primary anti-establishment. They favored Sanders in the Democratic primary.
That was my problem with Pacific Rim. I kept thinking, "those streets aren't engineered to take that kind of ground pressure."
As for the second shot in the clip linked above, I didn't notice it On closer viewing, yeah, it looks fake, but not really any faker than what Hollywood has been doing for decades with squibs, not to mention its very unrealistic but dramatically satisfying explosions. And it's not just Hollywood.
Here's the thing about 2016 -- it was a very close thing. Just 1/2 of 1% of the turnout in three states (Pennsylvania, Michigan, Wisconsin) would have flipped those states. That's 78,000 strategically based voters -- not even 1/10 of 1% of the total US votes cast -- and the Electoral College would have gone the other way.
The flaw in nearly every 2016 postmortem analysis I've seen is that they all look for the explanation. It's a situation tailor made for advancing pet theories: an idea that has any truth at all in it can quite plausibly be claimed to have flipped the results.
So you can't rule out Russian meddling by citing Clinton's (unquestioned) weaknesses as a candidate. They could *both* have been decisive.
From a financial perspective, this isn't a problem, unless your plan is to ditch every other form of transportation in favor of Hyperloop.
If you have 2000 passengers/hour who *want* to ride Hyperloop at the price you're charging, and you can only take 840/hour, you simply raise your prices until demand equals supply. It's basic economics.
You're required to pass a test on how to recognize fake ids, determine if someone has had too much to drink and needs to be shut off, and what your legal responsibilities and liabilities are as a server. The permit cost is $8.99, and includes a video tutorial.
That seems pretty reasonable to me. It's not like they're testing you on whether you can mix a Martini.
Well, consider the 67.3m Grenfell Tower which burned spectacularly last June, killing seventy people. The tower was reinforced concrete, but it was the decorative polyethylene cladding that transmitted the fire at deadly speed, and the interior apartment furnishings that actually killed people.
So it's quite possible for a concrete building to become a fire trap; it's the superficial bits that are the risk. Massive wooden structural members might burn in theory, but like an over-large log they wouldn't catch fire quickly.
So I should think that a large wooden building could in principle be engineered to be for all practical purposes as fire safe as concrete building. The problem is knowing that something is safe in practice. Engineering is as much about the application of experience as it is induction from general principles. So if you build far beyond the limits of experience, you can never be quite certain of the behavior of a system.
Of course I'm aware that the word "stolen" in this context isn't the same as someone "stealing" your bike. "Misappropriated" if you prefer, but trying to get the world onboard with your preferred word usage is like trying to nail jello to the wall, even if it's the very best jello.
If you've ever worked in technology and actually thought about what you were doing, you'd realize that you're building on the ideas of others.
If the only thing China were capable of doing is copying US tech, then US tech companies and the US military would have no real worries from Chinese tech espionage. By the time they got the tech working, we'd be onto the next big thing.
But China is a lot more capable than that. They're a huge country sitting on a huge talent pool with a regime that understands the value of technological research. When they steal US tech secrets they aren't just stealing grist for their mill; they're stealing seed corn.
Actually, this would make sense. In fact it's kind of the whole point.
As a private association, the FreeBSD developers are free to adopt whatever code of conduct they want. They could even adopt a developer dress code if they wanted to, and they wouldn't be oppressing anyone because if you didn't want to wear a straw boater hat and a bow tie you could take your contributed code -- an everyone else's code -- and start your own developer group.
In fact the only way to reconcile the freedom of a developer group to govern itself with the freedom of individuals to make their own choices is if the software itself is also free.
Do these microplastics affect the fish's health and / or significantly impact their lifespan? If not then is there some other reason to be concerned?
Well, the problem is with science itself. Or rather with the way science is covered in the media. The media wants answers, but the first step in science is finding good questions. You can't answer a question like that until you know the phenomenon exists, but people want to jump straight to what it means.
If there were one thing I wish the educational system instilled in people, it would be the capacity of being concerned without necessarily being alarmed. People come out with basically two easy options to fall back on: alarmism and denialism.
To put that into perspective, the US uses 6.5x as much coal on a per capita basis as Germany. So reductions in coal usage for them is bound to be harder because they're starting from a lower point, and they still have to provide baseload electricity generation.
Germany has recently experienced negative electricity prices in which consumers are paid to take power off the grid dumped in by renewables. So taking more coal plants offline and putting in more renewables is going to make that problem worse. They need either an alternative baseload generation source, or more energy storage.
If Russia is making the effort to propagandize the American public, that's prima facie evidence that they at least believe it is reasonably possible that it has an effect.
As to whether their efforts have borne fruit, I don't think we're in a position to say definitively that the burden of proof lies on one side or another. It's quite reasonable to take cautious precautionary steps, which isn't tantamount to pushing the panic button.
This is a strain known for producing severe flu cases and sometimes-fatal secondary complications.
There is evidence that in past A/H3N2 outbreaks, people who contracted the flu despite being vaccinated had less severe symptoms and fewer complications. Even if the vaccine is only "10% effective" at preventing infection, the evidence still suggests that it's worth getting, especially as this flu is claiming the lives of many young, healthy people.
IMPORTANT: the "10%" figure was an early estimate from Australia in December. More recent figures I've heard are 17% effectiveness and 30% effectiveness in the US.
In WW2 the British built a multi-role military aircraft called the "Mosquito" out of plywood. Originally conceived as a very fast lightweight bomber, it was a brilliant success at a wide variety of tasks: night fighter, high altitude interceptor, ground attack craft, photo-reconnaissance craft, torpedo bomber.
Basically it was the anti-F35: designed to do one thing well, it ended up doing everything pretty well.
Anyhow, what you're saying black faces have less luminance -- well sure. That says nothing about contrast. The idea that there's less detail there to be seen is a result of looking at poorly lit photographs.
The upshot is that if the problem is in the sensor's ability to handle a certain luminance range, by that theory the algorithms should perform better on black faces in very brightly lit conditions.
Also known as the base rate fallacy. If you're looking for a needle in a haystack, an algorithm which correctly distinguishes them 99% of the time is useless.
Slashdot Mirror
Well, dashing off a memo (one way or the other) doesn't take a lot of time.
What does take a lot of time, cumulatively speaking is reading such memos. If you send a memo and it gets distributed to a couple of hundred people, the cost of that memo can be staggering.
This is a "it depends" question. If you wrap your device in aluminum foil it can become capacitively coupled to the foil. You could fix that by putting it in a room completely lined with foil (increasing the thickness of the dialectric). Or you can connect your foil to a massive electron sink, like a large hunk of metal or the Earth.
Alternatively they could build a sun dial out of masonry, which is a cheap and long-lived material. That would keep time (during the day) as long as the structure existed.
Build it big enough and it could be quite precise, like this one.
As a numbers-oriented guy, it pains me to say this, but Clinton relied on the numbers *too* much.
All polls are adjusted by some kind of likely-voter model; all the "margin of error" figures you hear only take into account *random sampling* errors, not systematic ones. As useful as empirical numbers are, you have to keep in mind that they're only good as the assumptions under which they were collected.
This means that when Clinton got word of something happening that wasn't reflected in her internal polling, she should have taken it seriously rather than discounting it. It was incompetent, but in the opposite way that people are numerically incompetent. People usually trust their gut and discount hard data. But it's possible to trust hard data too much too.
By the way I don't have any problem with Russians or anyone else weighing in on our election -- as long as they do it above board. If they can, as Russians, sway American opinions, more power to them. Pretending to be Americans, to the point of stealing American identities is a different matter.
Again let me reiterate, the goal of Russian efforts seem to have been to undermine Americans' faith in their own country. It's quite possible they had no *intent* of swaying the election, what they may have wanted to do is inflame opposition to a Clinton presidency. Russia has a complicated relationship to the west, because Russia is a country run by economic parasites. Those parasite need stable democratic countries as place to store the wealth they've looted from their own country. Russian oligarchs are behind the insane real estate prices in places like New York, London, Vancouver, and Miami.
Yes, turnout is where the big swings happen in a country where about 40% of the electorate sit out each election; it's not always the same 40%.
Rural turnout was a big factor in 2016, and this speaks to the strengths and weaknesses of the candidates. Clinton, unlike her husband, wasn't good at projecting empathy to voters in rural districts ravaged by economic decline and the opioid crisis. Neither was Trump, but he was much better at projecting their *anger*.
This also shows why Russian meddling may have been quite effective. A lot of it is aimed at spreading disillusionment; it's primary anti-establishment. They favored Sanders in the Democratic primary.
That was my problem with Pacific Rim. I kept thinking, "those streets aren't engineered to take that kind of ground pressure."
As for the second shot in the clip linked above, I didn't notice it On closer viewing, yeah, it looks fake, but not really any faker than what Hollywood has been doing for decades with squibs, not to mention its very unrealistic but dramatically satisfying explosions. And it's not just Hollywood.
Here's the thing about 2016 -- it was a very close thing. Just 1/2 of 1% of the turnout in three states (Pennsylvania, Michigan, Wisconsin) would have flipped those states. That's 78,000 strategically based voters -- not even 1/10 of 1% of the total US votes cast -- and the Electoral College would have gone the other way.
The flaw in nearly every 2016 postmortem analysis I've seen is that they all look for the explanation. It's a situation tailor made for advancing pet theories: an idea that has any truth at all in it can quite plausibly be claimed to have flipped the results.
So you can't rule out Russian meddling by citing Clinton's (unquestioned) weaknesses as a candidate. They could *both* have been decisive.
They still haven't found the underground city of Atlantis!
Maybe because they keep looking in the ocean.
From a financial perspective, this isn't a problem, unless your plan is to ditch every other form of transportation in favor of Hyperloop.
If you have 2000 passengers/hour who *want* to ride Hyperloop at the price you're charging, and you can only take 840/hour, you simply raise your prices until demand equals supply. It's basic economics.
You're required to pass a test on how to recognize fake ids, determine if someone has had too much to drink and needs to be shut off, and what your legal responsibilities and liabilities are as a server. The permit cost is $8.99, and includes a video tutorial.
That seems pretty reasonable to me. It's not like they're testing you on whether you can mix a Martini.
Well, consider the 67.3m Grenfell Tower which burned spectacularly last June, killing seventy people. The tower was reinforced concrete, but it was the decorative polyethylene cladding that transmitted the fire at deadly speed, and the interior apartment furnishings that actually killed people.
So it's quite possible for a concrete building to become a fire trap; it's the superficial bits that are the risk. Massive wooden structural members might burn in theory, but like an over-large log they wouldn't catch fire quickly.
So I should think that a large wooden building could in principle be engineered to be for all practical purposes as fire safe as concrete building. The problem is knowing that something is safe in practice. Engineering is as much about the application of experience as it is induction from general principles. So if you build far beyond the limits of experience, you can never be quite certain of the behavior of a system.
Of course I'm aware that the word "stolen" in this context isn't the same as someone "stealing" your bike. "Misappropriated" if you prefer, but trying to get the world onboard with your preferred word usage is like trying to nail jello to the wall, even if it's the very best jello.
If you've ever worked in technology and actually thought about what you were doing, you'd realize that you're building on the ideas of others.
If the only thing China were capable of doing is copying US tech, then US tech companies and the US military would have no real worries from Chinese tech espionage. By the time they got the tech working, we'd be onto the next big thing.
But China is a lot more capable than that. They're a huge country sitting on a huge talent pool with a regime that understands the value of technological research. When they steal US tech secrets they aren't just stealing grist for their mill; they're stealing seed corn.
Actually, this would make sense. In fact it's kind of the whole point.
As a private association, the FreeBSD developers are free to adopt whatever code of conduct they want. They could even adopt a developer dress code if they wanted to, and they wouldn't be oppressing anyone because if you didn't want to wear a straw boater hat and a bow tie you could take your contributed code -- an everyone else's code -- and start your own developer group.
In fact the only way to reconcile the freedom of a developer group to govern itself with the freedom of individuals to make their own choices is if the software itself is also free.
Do these microplastics affect the fish's health and / or significantly impact their lifespan? If not then is there some other reason to be concerned?
Well, the problem is with science itself. Or rather with the way science is covered in the media. The media wants answers, but the first step in science is finding good questions. You can't answer a question like that until you know the phenomenon exists, but people want to jump straight to what it means.
If there were one thing I wish the educational system instilled in people, it would be the capacity of being concerned without necessarily being alarmed. People come out with basically two easy options to fall back on: alarmism and denialism.
Come back to my place and we'll put on some White Stripes.
To put that into perspective, the US uses 6.5x as much coal on a per capita basis as Germany. So reductions in coal usage for them is bound to be harder because they're starting from a lower point, and they still have to provide baseload electricity generation.
Germany has recently experienced negative electricity prices in which consumers are paid to take power off the grid dumped in by renewables. So taking more coal plants offline and putting in more renewables is going to make that problem worse. They need either an alternative baseload generation source, or more energy storage.
If you like it, you could pretend you don't understand.
If Russia is making the effort to propagandize the American public, that's prima facie evidence that they at least believe it is reasonably possible that it has an effect.
As to whether their efforts have borne fruit, I don't think we're in a position to say definitively that the burden of proof lies on one side or another. It's quite reasonable to take cautious precautionary steps, which isn't tantamount to pushing the panic button.
This is a strain known for producing severe flu cases and sometimes-fatal secondary complications.
There is evidence that in past A/H3N2 outbreaks, people who contracted the flu despite being vaccinated had less severe symptoms and fewer complications. Even if the vaccine is only "10% effective" at preventing infection, the evidence still suggests that it's worth getting, especially as this flu is claiming the lives of many young, healthy people.
IMPORTANT: the "10%" figure was an early estimate from Australia in December. More recent figures I've heard are 17% effectiveness and 30% effectiveness in the US.
In WW2 the British built a multi-role military aircraft called the "Mosquito" out of plywood. Originally conceived as a very fast lightweight bomber, it was a brilliant success at a wide variety of tasks: night fighter, high altitude interceptor, ground attack craft, photo-reconnaissance craft, torpedo bomber.
Basically it was the anti-F35: designed to do one thing well, it ended up doing everything pretty well.
Actually it is on point with regard to testing.
Anyhow, what you're saying black faces have less luminance -- well sure. That says nothing about contrast. The idea that there's less detail there to be seen is a result of looking at poorly lit photographs.
The upshot is that if the problem is in the sensor's ability to handle a certain luminance range, by that theory the algorithms should perform better on black faces in very brightly lit conditions.
No, it will find "needles" all over the place, laced through the haystack. Just one of those tens of thousands of "needles" will actually be a needle.
Also known as the base rate fallacy. If you're looking for a needle in a haystack, an algorithm which correctly distinguishes them 99% of the time is useless.