Interesting, but if it were from South America, shouldn't there be physical evidence in the manuscript itself, such as the makeup of the paper, that would identify it as originating there? After all, the manuscript has been dated to before Columbus' voyage, so it couldn't very well be a copy of New World writing. It had to either originate in Europe, and have no influence from the New World, or originate in the New World.
Not really. The new maps is basically built around this feature. This seems to me to be a case of people not understanding that "search nearby" is effectively the default Google Maps behavior in the new version. It only moves the map if it doesn't find nearby results, or if the search query is a specific location elsewhere.
Try it yourself. Using the new maps, zoom in on a location with a number of restaurants, and type "restaurants" in the search box.
Counter-evidence would be, say, a lack of increase in wildfires, a lack of increase in the frequency of very strong hurricanes, a lack of increase in storm activity in general, a lack of increase in measured average global temperatures.
Or it's that you're not understanding the science. Certainly there is no claim whatsoever that global warming causes all weather to become extreme. The claim is rather that the number of extreme events is increased by global warming, and furthermore that some events are so extreme that it is highly unlikely that they would have happened without global warming. Sandy was one of those events. Sandy could have happened without global warming, it's just unlikely (most likely the warmer ocean allowed the hurricane to both travel further north and remain stronger as it traveled).
You don't understand what quantitative easing is, do you?
Quantitative easing is the central bank buying certain securities (long-term government bonds and mortgage-backed securities). It buys less than one percent of the outstanding value of these securities each month. This isn't an "attack" on the functioning of the US dollar. It isn't even close. It's just an attempt to keep inflation stable at the margins by slightly reducing long-term interest rates.
Also, the bank bailouts aren't even remotely related to a 51% attack. They are problematic for entirely different reasons (mostly moral hazard), and Bitcoin is certainly not immune to the exact same kind of problem (if Bitcoin was ever trusted enough for people to start lending Bitcoins, that is).
Bullcrap. Bitcoin by design has an extremely volatile value. This makes it absolute crap as a currency. The entire benefit of having a currency is having something I can own which I can exchange for a known quantity of goods sometime down the road. Ideally, I won't have to worry too much whether I head to the store today or tomorrow: I should be able to buy pretty much the same amount of stuff. Bitcoin doesn't have that luxury: the value of Bitcoins has dropped by nearly half in a couple of days. That is not acceptable, and pretty much no store is going to ever publish the price of a product in Bitcoins as a result.
The author of this article basically just waves this point away because, "Economists can’t even agree on basic assumptions, which is why they argue endlessly," which is complete bullshit. There really isn't any significant argument among economists that we really should return to a gold standard (which, economically, is pretty much the same thing as Bitcoin, except not quite as bad because Bitcoins can be realistically traded at much higher rates). That argument is promoted pretty much solely by Libertarian cranks.
The article is lying out its ass, and the lie is incredibly transparent. To see that this is a lie, here is the actual SEC filing. It is gigantic, but the relevant bits are pages 358 and 359 of the PDF. There are three big things that become immediately obvious from a cursory analysis of the table and associated description:
The SEC is proposing a requirement that crowdfunders produce and submit a series of forms. They estimate the cost of producing these forms at $10,060, assuming use of a lawyer at $400/hour. There's also the cost of yearly audits for crowdfunding greater than $100,000.
The SEC is estimating that crowdfunding intermediaries will extract fees between 5% and 15%. This is obviously not a cost that the SEC is imposing, but is a fee imposed by intermediaries whose amount the SEC is not regulating.
The cost estimates used middle values for each bracket ($50,000, $300,000, and $750,000). This means that the claim that it might possibly cost $30,000 for somebody raising $100,000 in a crowdfunding project is pure, unadulterated fiction. The additional cost for that is around $10,000, plus the intermediary fee, at $400/hour for a lawyer. There's the additional cost of yearly audits for crowd funding of $100,001 or greater, which is unfortunate, but it puts lie to the ridiculous graphs put out by Sherwood Neiss.
As others have already mentioned, this is also only applicable to crowdfunding where securities are given to the funders. For that situation, the new regulations are actually a tremendous cost savings, as usually this would require an IPO, which typically cost millions of dollars. So somehow the moron that wrote this article changed a dramatic cost savings (from millions of dollars down to tens of thousands, possibly less for small amounts of funding with a cheap lawyer), and turned it into a story of "bureaucratic overreach". Come on. This article should be retracted for gross inaccuracy.
Welfare, housing assistance, charity. It's rough, but the basics are provided for if you go out and get them.
Knowing a number of extremely poor people, this isn't even remotely true. Especially not of late, when we have the national and state governments cutting back so severely on various programs to help the poor. I have one friend, for example, who has precisely zero normal income (due to various family issues and an untreated disability). Yet she has an incredibly hard time getting any sort of aid, because she has no proof of income!
This is why we need an unconditional basic income, instead of all these stupid programs. The very large number of conditions set on the various programs to help the poor end up guaranteeing that many extremely poor people get left out, if only because they have a hard time supplying the paperwork to prove that they qualify. Poor people also very frequently have a hard time traveling any significant distance, meaning that if they live in rural America, traveling to the various government offices to apply for aid becomes a significant burden. Some of the required documents (e.g. birth certificate) also come along with charges that are difficult to cover.
There is no good reason for this. Nobody deserves to be left destitute on the street, so we should just guarantee a base level of income so that nobody has to. Get your Social Security card, and get your monthly check, end of story (paid for with a moderate hike in income taxes). That way almost nobody will fall through the cracks.
And the most awesome thing about a guaranteed basic income that is high enough: if it is high enough that not working becomes a viable option, then it will break the stranglehold that employers have over their employees. Employers will actually have to provide decent working conditions and/or pay, or they will quickly find themselves without employees.
Certainly Snowden's actions made it a little bit more difficult for the US government to undertake certain actions. As those actions themselves are harmful and illegal, cry me a river. Snowden may not have helped the US government per se, but he helped the citizens of the US, and people around the world as well.
Yes, but at least this time they mentioned software emulation. That wasn't even mentioned before, making it sound more like a dual-boot system to me. Maybe next time this news story is posted they'll get the description entirely right? Possibly even mentioning that this is talking about tablets, not laptops or PC's?
Read the article. This is for devices that are convertible between tablet and laptop configurations with the help of a detachable keyboard. They also mention that they use an emulation layer for Android support. This is entirely believable, as Android has a Java-like interface, such that emulation essentially means building a Java-like virtual machine that has good performance. It's also not that big of a blow to MS, as MS still gets to sell the OS.
Not really. Just pointing out that one of the judge's justifications for finding this "legal" is absolutely bogus: that it's useful. It is absolutely not useful at all in preventing terror attacks. It might be useful in helping to pinpoint perpetrators after the fact, but it's worthless as a preventative measure. And, of course, as many others have noted, how useful it is is independent of its unconstitutionality. One wonders, then, why the judge even mentioned that.
Take a look at the article. It isn't just that nothing happened, but they didn't even have any positive detections. And it only takes a little bit of knowledge of neural networks and statistics to understand that this kind of program is doomed from the very start: the signal these data collection systems are looking for is exceedingly tiny. There are at most dozens of actual terrorists seriously working on plans to attack somewhere in the US at any given moment, compared to a population of over 300 million. In order for the system to actually detect those terrorists, then, it needs to have a detection accuracy that is better than the ratio of the non-terrorists to the total population. Otherwise, most of the "detections" will actually be false detections that do nothing but mislead law enforcement and infringe on the rights of innocent citizens. If we assume 50 terrorists, that means the system would need a precision better than 99.99998%.
It's almost impossible for any learning model to have a precision that high. Learning models in general have this problem with what are called long-tail errors. If you want to know what these errors are, check out Google's or Apple's speech recognition software on a smartphone (you can also access Google's speech recognition on the web on any PC with a microphone). Most of the time, it's pretty good. But sometimes the mis-detections are so far off it's ludicrous. Why in the world would you ever trust a system that needs obscenely high precision to an algorithm that has such nasty errors?
To be fair, it is possible to reduce the needed precision by filtering out people in the population who are unlikely to be terrorists (e.g. children, the elderly, women), but it's just not possible to reduce it enough to make sifting through large pools of information worthwhile.
Even if we disregard the obviously nasty privacy implications, in what way is a completely and utterly ineffective program "valuable"? I mean, come on. This extremely expensive program has stopped precisely zero attacks (source). I seriously hope the ACLU's lawyers are up to the task of arguing the idiocy of this program.
You do realize that there are cheaper and more energy-efficient methods to heat a home (or anything else) than incandescent bulbs, right? Like 2-5 times more efficient?
True. But for the vast majority of applications, that greater number of registers only translates into a small performance increase. I can potentially see x32 being useful for a rather small amount of heavily hand-optimized code (e.g. a massively optimized math or physics library), but for the vast majority of applications this performance benefit will be tiny.
To me, the real problem for the adoption of x32 is that so few programs on PC's need to worry that much about optimization. When it does become worthwhile for them to worry about optimization, there are likely to be many things that are more worthwhile to tackle for improving performance (e.g. algorithmic inefficiencies, using excessive I/O).
Slashdot Mirror
Interesting, but if it were from South America, shouldn't there be physical evidence in the manuscript itself, such as the makeup of the paper, that would identify it as originating there? After all, the manuscript has been dated to before Columbus' voyage, so it couldn't very well be a copy of New World writing. It had to either originate in Europe, and have no influence from the New World, or originate in the New World.
Not really. The new maps is basically built around this feature. This seems to me to be a case of people not understanding that "search nearby" is effectively the default Google Maps behavior in the new version. It only moves the map if it doesn't find nearby results, or if the search query is a specific location elsewhere.
Try it yourself. Using the new maps, zoom in on a location with a number of restaurants, and type "restaurants" in the search box.
Sure. The claim that extreme weather events are increasing due to climate change is falsifiable. What's your point?
Counter-evidence would be, say, a lack of increase in wildfires, a lack of increase in the frequency of very strong hurricanes, a lack of increase in storm activity in general, a lack of increase in measured average global temperatures.
Or it's that you're not understanding the science. Certainly there is no claim whatsoever that global warming causes all weather to become extreme. The claim is rather that the number of extreme events is increased by global warming, and furthermore that some events are so extreme that it is highly unlikely that they would have happened without global warming. Sandy was one of those events. Sandy could have happened without global warming, it's just unlikely (most likely the warmer ocean allowed the hurricane to both travel further north and remain stronger as it traveled).
Then this change doesn't impact you. If you don't have a Google+ profile, then obviously nobody can send e-mails to your Google+ profile.
It actually includes better spam controls. Seems like a pretty good system to me.
You don't understand what quantitative easing is, do you?
Quantitative easing is the central bank buying certain securities (long-term government bonds and mortgage-backed securities). It buys less than one percent of the outstanding value of these securities each month. This isn't an "attack" on the functioning of the US dollar. It isn't even close. It's just an attempt to keep inflation stable at the margins by slightly reducing long-term interest rates.
Also, the bank bailouts aren't even remotely related to a 51% attack. They are problematic for entirely different reasons (mostly moral hazard), and Bitcoin is certainly not immune to the exact same kind of problem (if Bitcoin was ever trusted enough for people to start lending Bitcoins, that is).
Bullcrap. Bitcoin by design has an extremely volatile value. This makes it absolute crap as a currency. The entire benefit of having a currency is having something I can own which I can exchange for a known quantity of goods sometime down the road. Ideally, I won't have to worry too much whether I head to the store today or tomorrow: I should be able to buy pretty much the same amount of stuff. Bitcoin doesn't have that luxury: the value of Bitcoins has dropped by nearly half in a couple of days. That is not acceptable, and pretty much no store is going to ever publish the price of a product in Bitcoins as a result.
The author of this article basically just waves this point away because, "Economists can’t even agree on basic assumptions, which is why they argue endlessly," which is complete bullshit. There really isn't any significant argument among economists that we really should return to a gold standard (which, economically, is pretty much the same thing as Bitcoin, except not quite as bad because Bitcoins can be realistically traded at much higher rates). That argument is promoted pretty much solely by Libertarian cranks.
The article is lying out its ass, and the lie is incredibly transparent. To see that this is a lie, here is the actual SEC filing. It is gigantic, but the relevant bits are pages 358 and 359 of the PDF. There are three big things that become immediately obvious from a cursory analysis of the table and associated description:
As others have already mentioned, this is also only applicable to crowdfunding where securities are given to the funders. For that situation, the new regulations are actually a tremendous cost savings, as usually this would require an IPO, which typically cost millions of dollars. So somehow the moron that wrote this article changed a dramatic cost savings (from millions of dollars down to tens of thousands, possibly less for small amounts of funding with a cheap lawyer), and turned it into a story of "bureaucratic overreach". Come on. This article should be retracted for gross inaccuracy.
Welfare, housing assistance, charity. It's rough, but the basics are provided for if you go out and get them.
Knowing a number of extremely poor people, this isn't even remotely true. Especially not of late, when we have the national and state governments cutting back so severely on various programs to help the poor. I have one friend, for example, who has precisely zero normal income (due to various family issues and an untreated disability). Yet she has an incredibly hard time getting any sort of aid, because she has no proof of income!
This is why we need an unconditional basic income, instead of all these stupid programs. The very large number of conditions set on the various programs to help the poor end up guaranteeing that many extremely poor people get left out, if only because they have a hard time supplying the paperwork to prove that they qualify. Poor people also very frequently have a hard time traveling any significant distance, meaning that if they live in rural America, traveling to the various government offices to apply for aid becomes a significant burden. Some of the required documents (e.g. birth certificate) also come along with charges that are difficult to cover.
There is no good reason for this. Nobody deserves to be left destitute on the street, so we should just guarantee a base level of income so that nobody has to. Get your Social Security card, and get your monthly check, end of story (paid for with a moderate hike in income taxes). That way almost nobody will fall through the cracks.
And the most awesome thing about a guaranteed basic income that is high enough: if it is high enough that not working becomes a viable option, then it will break the stranglehold that employers have over their employees. Employers will actually have to provide decent working conditions and/or pay, or they will quickly find themselves without employees.
Certainly Snowden's actions made it a little bit more difficult for the US government to undertake certain actions. As those actions themselves are harmful and illegal, cry me a river. Snowden may not have helped the US government per se, but he helped the citizens of the US, and people around the world as well.
Thank the person that brought these security breaches to light, not the people who have been illegally performing them.
Though technically, if you wanted to do this, you could just download the Android cross-platform SDK and use the emulator that comes with that.
Yes, but at least this time they mentioned software emulation. That wasn't even mentioned before, making it sound more like a dual-boot system to me. Maybe next time this news story is posted they'll get the description entirely right? Possibly even mentioning that this is talking about tablets, not laptops or PC's?
So, what does this offer that Docs in Google Drive doesn't?
Hey, at least it's a better description this time.
Read the article. This is for devices that are convertible between tablet and laptop configurations with the help of a detachable keyboard. They also mention that they use an emulation layer for Android support. This is entirely believable, as Android has a Java-like interface, such that emulation essentially means building a Java-like virtual machine that has good performance. It's also not that big of a blow to MS, as MS still gets to sell the OS.
Not really. Just pointing out that one of the judge's justifications for finding this "legal" is absolutely bogus: that it's useful. It is absolutely not useful at all in preventing terror attacks. It might be useful in helping to pinpoint perpetrators after the fact, but it's worthless as a preventative measure. And, of course, as many others have noted, how useful it is is independent of its unconstitutionality. One wonders, then, why the judge even mentioned that.
Yup. Pretty much why the entire program is unworkable.
Take a look at the article. It isn't just that nothing happened, but they didn't even have any positive detections. And it only takes a little bit of knowledge of neural networks and statistics to understand that this kind of program is doomed from the very start: the signal these data collection systems are looking for is exceedingly tiny. There are at most dozens of actual terrorists seriously working on plans to attack somewhere in the US at any given moment, compared to a population of over 300 million. In order for the system to actually detect those terrorists, then, it needs to have a detection accuracy that is better than the ratio of the non-terrorists to the total population. Otherwise, most of the "detections" will actually be false detections that do nothing but mislead law enforcement and infringe on the rights of innocent citizens. If we assume 50 terrorists, that means the system would need a precision better than 99.99998%.
It's almost impossible for any learning model to have a precision that high. Learning models in general have this problem with what are called long-tail errors. If you want to know what these errors are, check out Google's or Apple's speech recognition software on a smartphone (you can also access Google's speech recognition on the web on any PC with a microphone). Most of the time, it's pretty good. But sometimes the mis-detections are so far off it's ludicrous. Why in the world would you ever trust a system that needs obscenely high precision to an algorithm that has such nasty errors?
To be fair, it is possible to reduce the needed precision by filtering out people in the population who are unlikely to be terrorists (e.g. children, the elderly, women), but it's just not possible to reduce it enough to make sifting through large pools of information worthwhile.
Even if we disregard the obviously nasty privacy implications, in what way is a completely and utterly ineffective program "valuable"? I mean, come on. This extremely expensive program has stopped precisely zero attacks (source). I seriously hope the ACLU's lawyers are up to the task of arguing the idiocy of this program.
So, who gives a shit how many people die because of increased storms and rising sea levels? Nice.
You do realize that there are cheaper and more energy-efficient methods to heat a home (or anything else) than incandescent bulbs, right? Like 2-5 times more efficient?
True. But for the vast majority of applications, that greater number of registers only translates into a small performance increase. I can potentially see x32 being useful for a rather small amount of heavily hand-optimized code (e.g. a massively optimized math or physics library), but for the vast majority of applications this performance benefit will be tiny.
To me, the real problem for the adoption of x32 is that so few programs on PC's need to worry that much about optimization. When it does become worthwhile for them to worry about optimization, there are likely to be many things that are more worthwhile to tackle for improving performance (e.g. algorithmic inefficiencies, using excessive I/O).