You're right that OCR / voice recognition wouldn't be an interesting hurdle at all. Decent OCR (especially since you know Jeopardy uses clear text and a well-known font) is a solved problem. You might as well give the question to Watson in text, but with a delay equal to how long a human takes to read. (Most Jeopardy players simply read the question and use the time Trebek is talking to work on the answer.) Except that then the problem Watson is trying to solve changes from "determine the answer to the question and confidence level within N seconds" to "...within (N-K) seconds". If your time to solve the problem is reduced from 3 seconds to 2, you can probably overcome this with 50% more hardware. Throwing even twice as much hardware at the problem is neither restrictive nor at all interesting.
One of the major sources of security bugs in Java is that the Java standard libraries are primarily native code. The native code is rife with buffer overflows, unchecked pointers, and other vulnerabilities.
Both Android phones and iPhones have two sections of storage: one "internal" and one "external". The internal storage is not available as a USB mass storage device, only the external.
I don't know the details of the iPhone's system. On Android, nearly all applications and their data are stored on the internal storage that is not available over USB. You need to have a rooted phone in order to access data owned by applications on your phone. (The exceptions are apps that have been "installed to SD card", which is an option that the app can disable, and data that is intentionally stored on the SD card for the purpose of making it accessible over USB.)
Are you complaining that you didn't learn IT in college, or that you didn't learn how to research (the look-up kind), apply the results, and solve problems? If the latter, you should've gone for a better school or a more appropriate degree program.
Sure, many specific problems, particularly in fields like tech support, boil down to looking things up and not failing at it. Those people are more valuable when they don't have to look the things up in the first place, but you can get by with looking things up.
For people who build better wheels, Google is a poor answer.
The two "expanding into somethings" are different, though. You're imagining the 3D stuff of the Universe expanding into some empty 3D space that's surrounding it. This isn't the case with the 2D surface of a balloon -- it's "expanding" in its embedding space (the 3D space that holds the balloon's surface). It's not expanding into adjacent 2D space that was empty.
Unfortunately, the balloon analogy does have the 3D balloon expanding. This is not at all necessary. An infinite 2D space can fit inside a finite 3D space, you just have to fold it up a lot. Likewise, an infinite 3D space can fit into a finite 4D space. So the 3D universe can "expand" without there being "more space" at all.
To be pedantic, a 1D space with no beginning or end that has a finite size must be embedded in a 2D space. A 1D space with no beginning or end that has an infinite size can fit in a 1D space.
Well, if the Universe is a 3-space embedded in a 4-space (or modify as appropriate), then the "size" of the universe has a bit different meaning that what you're thinking, but is still meaningful.
Compare to a 2D space embedded in a 3D space. Usually the example is the inside surface of a balloon, since it shares some useful properties with the universe. It's equally valid to consider the surface of the Earth. To a person on Earth considering its 2D "surface", there is no identifiable point or line that is the "end" of the Earth. Nonetheless, the Earth's surface has a finite area.
Likewise, the 3D universe can have a finite volume without having an edge.
You need a huge mass to bend light by any measurable amount, and large masses (larger than those of Earth) are hard to measure accurately. So you have to measure two hard-to-measure things (light curvature and a huge mass) in order to get your reference, meaning that your unit is low-precision.
It's impractical to measure. The mass of a C12 atom is directly based on Planck's constant, so if you're going to define a kg by the mass of a C12 atom but can't reliably get an exactly number of C12 atoms all in one place (which would make a good standard), then you might as well just define it by the controlling constant, which is exactly what they're trying to do.
It's important to have all base units be as precise as possible. This is essentially the argument against "just averaging" them -- you lose a bunch of precision, and that's precision that we want to keep.
A balance scale measures mass, provided the force of gravity is the same on both sides of the balance. Other types of scales, on the other hand, measure gravitational force.
They're linearly dependent, actually (in the sub-relativistic regime).
Don't go overboard from the correct statement of "mass and weight are not the same thing" to "mass and weight are unrelated", which is what "independent" in any scientific context means.
The mass of a cubic centimeter of water spans a relatively huge range, depending on the composition of the water and the circumstances under which it's measured. The composition of even distilled water varies, since both hydrogen and oxygen have a variety of isotopes, the ratios of which vary from one source of water to another.
If you boil this down to a theoretical, idealized system (e.g., using a composition of water that's impossible to reproduce), you might as well base it on something more stable, like the mass of a particular kind of atom. If you do that, you might as well simply base it on a fundamental constant, which is exactly what TFA is talking about. A fundamental constant is really the way to go here. The fact that different ways of measuring it disagree is really just a minor bump in the road.
And only relevant if you need Google to help finish words for you.
Interestingly, "t" through "tor" autocomplete (to words not including "torrent"). Once you've typed "torrent", you get multiple-word suggestions that include the word "torrent". If you add more letters, like "i" (the prefix to "torrential"), it returns to normal autocomplete behavior.
Slashdot Mirror
Although there is a Toronto in the U.S. and the Toronto in Canada has an American League baseball team.
You're right that OCR / voice recognition wouldn't be an interesting hurdle at all. Decent OCR (especially since you know Jeopardy uses clear text and a well-known font) is a solved problem. You might as well give the question to Watson in text, but with a delay equal to how long a human takes to read. (Most Jeopardy players simply read the question and use the time Trebek is talking to work on the answer.) Except that then the problem Watson is trying to solve changes from "determine the answer to the question and confidence level within N seconds" to "...within (N-K) seconds". If your time to solve the problem is reduced from 3 seconds to 2, you can probably overcome this with 50% more hardware. Throwing even twice as much hardware at the problem is neither restrictive nor at all interesting.
http://en.wikipedia.org/wiki/Ashkenazi
One of the major sources of security bugs in Java is that the Java standard libraries are primarily native code. The native code is rife with buffer overflows, unchecked pointers, and other vulnerabilities.
Both Android phones and iPhones have two sections of storage: one "internal" and one "external". The internal storage is not available as a USB mass storage device, only the external.
I don't know the details of the iPhone's system. On Android, nearly all applications and their data are stored on the internal storage that is not available over USB. You need to have a rooted phone in order to access data owned by applications on your phone. (The exceptions are apps that have been "installed to SD card", which is an option that the app can disable, and data that is intentionally stored on the SD card for the purpose of making it accessible over USB.)
Are you complaining that you didn't learn IT in college, or that you didn't learn how to research (the look-up kind), apply the results, and solve problems? If the latter, you should've gone for a better school or a more appropriate degree program.
Sure, many specific problems, particularly in fields like tech support, boil down to looking things up and not failing at it. Those people are more valuable when they don't have to look the things up in the first place, but you can get by with looking things up.
For people who build better wheels, Google is a poor answer.
The whole idea of penalizing people for using the internet to produce answers in today's world sounds silly now doesn't it?
Not if you want to hire them to perform a job more demanding than looking up information on the Internet.
The two "expanding into somethings" are different, though. You're imagining the 3D stuff of the Universe expanding into some empty 3D space that's surrounding it. This isn't the case with the 2D surface of a balloon -- it's "expanding" in its embedding space (the 3D space that holds the balloon's surface). It's not expanding into adjacent 2D space that was empty.
Unfortunately, the balloon analogy does have the 3D balloon expanding. This is not at all necessary. An infinite 2D space can fit inside a finite 3D space, you just have to fold it up a lot. Likewise, an infinite 3D space can fit into a finite 4D space. So the 3D universe can "expand" without there being "more space" at all.
To be pedantic, a 1D space with no beginning or end that has a finite size must be embedded in a 2D space. A 1D space with no beginning or end that has an infinite size can fit in a 1D space.
Well, if the Universe is a 3-space embedded in a 4-space (or modify as appropriate), then the "size" of the universe has a bit different meaning that what you're thinking, but is still meaningful.
Compare to a 2D space embedded in a 3D space. Usually the example is the inside surface of a balloon, since it shares some useful properties with the universe. It's equally valid to consider the surface of the Earth. To a person on Earth considering its 2D "surface", there is no identifiable point or line that is the "end" of the Earth. Nonetheless, the Earth's surface has a finite area.
Likewise, the 3D universe can have a finite volume without having an edge.
You need a huge mass to bend light by any measurable amount, and large masses (larger than those of Earth) are hard to measure accurately. So you have to measure two hard-to-measure things (light curvature and a huge mass) in order to get your reference, meaning that your unit is low-precision.
Silver oxidizes. Platinum-iridium alloy is more stable than gold.
That's exactly what's suggested in the article.
The problem is that you have no appreciate of the time different for "shoving a few atoms around" and counting one kg of atoms.
It's impractical to measure. The mass of a C12 atom is directly based on Planck's constant, so if you're going to define a kg by the mass of a C12 atom but can't reliably get an exactly number of C12 atoms all in one place (which would make a good standard), then you might as well just define it by the controlling constant, which is exactly what they're trying to do.
It's important to have all base units be as precise as possible. This is essentially the argument against "just averaging" them -- you lose a bunch of precision, and that's precision that we want to keep.
A balance scale measures mass, provided the force of gravity is the same on both sides of the balance. Other types of scales, on the other hand, measure gravitational force.
UTC is a terrible time reference: it also includes leap seconds, which are more or less arbitrary.
Mass and weight are independent.
They're linearly dependent, actually (in the sub-relativistic regime).
Don't go overboard from the correct statement of "mass and weight are not the same thing" to "mass and weight are unrelated", which is what "independent" in any scientific context means.
Circular definitions aren't really a problem for anyone well-versed in algebra.
The mass of a cubic centimeter of water spans a relatively huge range, depending on the composition of the water and the circumstances under which it's measured. The composition of even distilled water varies, since both hydrogen and oxygen have a variety of isotopes, the ratios of which vary from one source of water to another.
If you boil this down to a theoretical, idealized system (e.g., using a composition of water that's impossible to reproduce), you might as well base it on something more stable, like the mass of a particular kind of atom. If you do that, you might as well simply base it on a fundamental constant, which is exactly what TFA is talking about. A fundamental constant is really the way to go here. The fact that different ways of measuring it disagree is really just a minor bump in the road.
So, there's no such thing as genes that control particular behaviors that you define, but there are genes that control other behaviors.
I'm pretty sure the only way to settle this is empirically.
And only relevant if you need Google to help finish words for you.
Interestingly, "t" through "tor" autocomplete (to words not including "torrent"). Once you've typed "torrent", you get multiple-word suggestions that include the word "torrent". If you add more letters, like "i" (the prefix to "torrential"), it returns to normal autocomplete behavior.
Perhaps the new standard will be to prefix with a z, so that you can autocomplete zbittorrent.
All they're asking for is for ISPs to retain DHCP logs longer.