You're right - for predictions, you should be looking at fellow megaguru Kurzweil!
And don't be so impressed by Wolfram's Alpha - at least last week when you asked it to evaluate 20000! (that's factorial), it would give an answer that was out by a factor of 20000! (that's not factorial, that's an exclamation.) Dumbass out-by-one error - heads should roll for something as stupid as that (it's an easy mistake to make, but it's equally easy to test the answer's right - the lack of testing is the unforgivable thing). Can you imagine a mobile phone that failed to dial the final digit in a telephone number, or a word-processor which would always chop off your final character?
You don't have that option in the test they did. They took a gay and a straight, and you (the mathematical model) had to determine which one was gay and which was straight. You can't apply your method to this test.
^^^ Most informative post on the story so far. I was just about to post something similar myself, but you saved me the effort.
The middle link, to the PDF, does wave some statistical terminology in my face, but alas I don't have the stats nous to work out how interesting their claim is from is. Apparently, they give the probability of correctly working out which is which when given two people from opposite sides of a fence. Just knowing that you've got both sides of a dichotomy gives you a pretty useful dollop of information. When I've seen tests like that in the past, namely can you distinguish X from non-X, the subjects were given pairs that were X+X, X+nonX, and nonX+nonX. (See the Mythbusters episode that covered "smell of fear", IIRC, for such tests.)
It would be appreciated if someone with better stats smarts could chip in.
I'd like to get my hands on the raw data set and see if a heuristic algoritm can do better than their arithmetical one. (i.e. "obvious", as the story's tags suggest.)
Are you sure? Are you sure it's not just the unconstrained rambling of a man who's obsessed with a single subject (that might happen to be an interesting subject, but that's irrelevant)?
But there are synonyms that need to be covered. They're a commercial entity, so would want both amazon.commercial and amazon.commerce - I have no problem with that. Again, we could shorten both of those too.
Stop whining here about such trivial matters - if the guy has annoyed you, then take it up with his boss. You can make an appointment to speak to him any Sunday, (but don't be surprised if he doesn't return your calls).
The higher latitudes of the Northern Hemisphere are actually quite small. Large proportions of it are land, but it's just not that big. Try an area-preserving projection: http://upload.wikimedia.org/wikipedia/commons/6/67/Lambert_cylindrical_equal-area_projection_SW.jpg
I think originally the coochie snorcher (not that I know what that verb even means) was 13, not even 15. (And 15 wouldn't even be considered statutory rape in some places, the age of consent is 14 where I live, for example.) I seem to remember there being some outcry about how the material behind that piece was even researched - it could even be considered child abuse in itself. (So in response to the ACs, it is not just depicting child abuse in a work of art, it actually involved child abuse in its creation.)
The author was somewhere on the fucked-up spectrum, that's all I know. But I would say that, potential rapist that I am.
Pah, Ghostrider's irresponsible - you shouldn't be racing on public roads. This is far better: http://www.youtube.com/watch?v=Ph0SjLC_lqk because, erm, because, erm,...
>Canon's sensor technology has hit a plateau... >Nikon is using Sony sensors
So arguably, Canon's sensors, by existing, are doing better than Nikon's nonexistent sensors. Nikon also use Sanyo sensors in their cheap stuff, have done for years. I'm not saying there's anything wrong with just buying in better components than you could make yourself, of course.
>We can't extrapolate to any DSLR improvements yet.
The technology of the sensor is orthogonal to whether there's a piece of silvered glass flapping up and down in front of it. We can extrapolate to there being extremely high sensitivity 2MP DSLRs immediately, as all they have to do is stick their electronics in a conventional box. We can't extrapolate to any high resolution still image 35mm format photography improvements yet.
> but let's assume that "microns square" means square pixels 19 microns on each side
You mean "let's read English coreectly"? Well done, you must be very proud of yourself! An "$N $unit square" has always meant a square with an area of $N*$N square units.
> I happen to be a Nikon fanboy
Should we conclude anything about being a Nikon fanboi and an inability to understand simple concepts?
> Nikon beat Canon to market with a 36.3 million pixel full-frame sensor
Despite your stupidity, you must be very proud of your absolutely mastodonic penis.
It also eliminates huge amounts of stuff the eye can see. It also eliminates lost of stuff that we don't care about, such as small changes over time, with its motion estimation. Both DCT and ME are very flawed, but *good enough most of the time*.
Sensor noise isn't the noise problem. The fact that signal noise is proportional to the square root of the signal is the noise problem. This is statistics, not electronics.
In a (D)SLR there is a single light path with a mirror in the way. Moving the mirror out of the way does nothing but restore the single light path. There's nowhere to put a second sensor. However, most serious users of interchangeable lens cameras will have multiple bodies, so I just expect people to carry around another, high sensitivity, body in their bag. Presumably with their f/1.2 50mm on it, so they're ready to just start grabbing photons.
Bizarre. KAT has a very wide selection, and as long as you have javascript turned off it's a very clean and easily navigable site. Presently it's my favourite site for just about every reason imaginable.
Slashdot Mirror
I presume that the military/security advisors are as well informed in their field as the ecomonic ones are in their field. I.e. not at all.
You're right - for predictions, you should be looking at fellow megaguru Kurzweil!
And don't be so impressed by Wolfram's Alpha - at least last week when you asked it to evaluate 20000! (that's factorial), it would give an answer that was out by a factor of 20000! (that's not factorial, that's an exclamation.) Dumbass out-by-one error - heads should roll for something as stupid as that (it's an easy mistake to make, but it's equally easy to test the answer's right - the lack of testing is the unforgivable thing). Can you imagine a mobile phone that failed to dial the final digit in a telephone number, or a word-processor which would always chop off your final character?
You lose - you failed to even answer the test question.
The test question is "which of subjects A and B, of which one is straight and one is gay, is the gay one?".
Not quite - if it's a bear, then it's 88% likely to like shitting in the woods!
You don't have that option in the test they did. They took a gay and a straight, and you (the mathematical model) had to determine which one was gay and which was straight. You can't apply your method to this test.
If he's gay he could well be a closet Republican that's in denial.
^^^
Most informative post on the story so far. I was just about to post something similar myself, but you saved me the effort.
The middle link, to the PDF, does wave some statistical terminology in my face, but alas I don't have the stats nous to work out how interesting their claim is from is. Apparently, they give the probability of correctly working out which is which when given two people from opposite sides of a fence. Just knowing that you've got both sides of a dichotomy gives you a pretty useful dollop of information. When I've seen tests like that in the past, namely can you distinguish X from non-X, the subjects were given pairs that were X+X, X+nonX, and nonX+nonX. (See the Mythbusters episode that covered "smell of fear", IIRC, for such tests.)
It would be appreciated if someone with better stats smarts could chip in.
I'd like to get my hands on the raw data set and see if a heuristic algoritm can do better than their arithmetical one. (i.e. "obvious", as the story's tags suggest.)
Are you sure? Are you sure it's not just the unconstrained rambling of a man who's obsessed with a single subject (that might happen to be an interesting subject, but that's irrelevant)?
But there are synonyms that need to be covered. They're a commercial entity, so would want both amazon.commercial and amazon.commerce - I have no problem with that. Again, we could shorten both of those too.
I'll grab .load, and can do you a twofer on .butt.load and .shit.load?
Stop whining here about such trivial matters - if the guy has annoyed you, then take it up with his boss. You can make an appointment to speak to him any Sunday, (but don't be surprised if he doesn't return your calls).
The higher latitudes of the Northern Hemisphere are actually quite small. Large proportions of it are land, but it's just not that big. Try an area-preserving projection:
http://upload.wikimedia.org/wikipedia/commons/6/67/Lambert_cylindrical_equal-area_projection_SW.jpg
I think they do care whether it functions or not. They want it to not function.
Were Saturday and Sunday to be refered to as a pair of weekends, your analogy would be perfect.
I think originally the coochie snorcher (not that I know what that verb even means) was 13, not even 15. (And 15 wouldn't even be considered statutory rape in some places, the age of consent is 14 where I live, for example.) I seem to remember there being some outcry about how the material behind that piece was even researched - it could even be considered child abuse in itself. (So in response to the ACs, it is not just depicting child abuse in a work of art, it actually involved child abuse in its creation.)
The author was somewhere on the fucked-up spectrum, that's all I know. But I would say that, potential rapist that I am.
"230 to 295 degrees Fahrenheit"
I know that 1000 F was defined as the temperature of the king of france's lit fart, so it's about a quarter of the temperature of a lit fart.?
Pah, Ghostrider's irresponsible - you shouldn't be racing on public roads. ...
This is far better:
http://www.youtube.com/watch?v=Ph0SjLC_lqk
because, erm, because, erm,
If you throw in giant pandas too - I'm with you!
>Canon's sensor technology has hit a plateau ...
>Nikon is using Sony sensors
So arguably, Canon's sensors, by existing, are doing better than Nikon's nonexistent sensors. Nikon also use Sanyo sensors in their cheap stuff, have done for years. I'm not saying there's anything wrong with just buying in better components than you could make yourself, of course.
>We can't extrapolate to any DSLR improvements yet.
The technology of the sensor is orthogonal to whether there's a piece of silvered glass flapping up and down in front of it. We can extrapolate to there being extremely high sensitivity 2MP DSLRs immediately, as all they have to do is stick their electronics in a conventional box. We can't extrapolate to any high resolution still image 35mm format photography improvements yet.
> but let's assume that "microns square" means square pixels 19 microns on each side
You mean "let's read English coreectly"? Well done, you must be very proud of yourself! An "$N $unit square" has always meant a square with an area of $N*$N square units.
> I happen to be a Nikon fanboy
Should we conclude anything about being a Nikon fanboi and an inability to understand simple concepts?
> Nikon beat Canon to market with a 36.3 million pixel full-frame sensor
Despite your stupidity, you must be very proud of your absolutely mastodonic penis.
It also eliminates huge amounts of stuff the eye can see. It also eliminates lost of stuff that we don't care about, such as small changes over time, with its motion estimation. Both DCT and ME are very flawed, but *good enough most of the time*.
Sensor noise isn't the noise problem. The fact that signal noise is proportional to the square root of the signal is the noise problem. This is statistics, not electronics.
In a (D)SLR there is a single light path with a mirror in the way. Moving the mirror out of the way does nothing but restore the single light path. There's nowhere to put a second sensor. However, most serious users of interchangeable lens cameras will have multiple bodies, so I just expect people to carry around another, high sensitivity, body in their bag. Presumably with their f/1.2 50mm on it, so they're ready to just start grabbing photons.
Shutting down torrent-hosting sites does nothing to ameliorate their sales figures in such cases.
``no other source [being] readily available [...] is a significant source of "lost sales"''
The businesses are still failing to respond to an economic want. They have no right to survive, they've been terminally stupid for so long.
Bizarre. KAT has a very wide selection, and as long as you have javascript turned off it's a very clean and easily navigable site. Presently it's my favourite site for just about every reason imaginable.