That said, here's an algorithm I just made up that ten (metaphorical) bucks say will kick his GA's ass.
Undue bravado: easier said than done! I whipped up some code to do roughly what I described in my previous post, and though it works -- and perhaps more quickly than Alsling's GA, which took a million generations -- the results are much, much less impressive: It doesn't look as good, and it uses more polygons (however, I use triangles). I think a lot of it has to do with the fact that, although I can achieve a reasonable square error with my method, it neglects the low frequencies, which is actually what the eye pays attention to.
First, a disclaimer, because I'm a nice guy: What this guy did is a neat sort of hack, and I approve. Many of us could have done this just as well, but this guy bothered and we didn't, so he deserves credit, and we should be happy for and with him. And it looks cool; that counts for a lot. So let us not rain on his parade. Sincere kudos are in order.
That said, here's an algorithm I just made up that ten (metaphorical) bucks say will kick his GA's ass. It stems from the fact that he's blending the polygons, which is, if it's additive blending he's using, a linear operation (if not, you can probably perform a transformation on the color values to turn it into something linear; e.g., take a log if it's multiplicative blending.)
Step 1: Generate a bajillion random images of polygons.
Step 2: Perform Graham-Schmidt orthonormalization on said images, while keeping track of the coefficients. In other words, do QR factorization.
Step 3: Project the Mona Lisa onto these vectors. I.e., premultiply it by Q^T.
Step 4: Throw out all of components of the projected Mona Lisa that fall below some threshold.
Step 5: Generate polygon-a-Lisa as a linear combination of images (from step 2) using the coefficients you computed in step 3 and kept in step 4.
The end! Isn't linear algebra fun?!:-) And of course you can rearrange the steps and do some similar variations of this to make the whole process less memory intensive; you don't need to actually store all these random images; you can instead do the projection and thresholding as you generate random polygons.
What's frightening about this particular case is that I actually know about these files they're referring to -- and tons of people used to have them. Back when DirectConnect was a popular P2P app, I remember seeing files with names like "Simpsons porn" on lots of people's hard drives. I never thought much of it; obviously it was a crude joke and nothing more. It would never have even occurred to me to consider it child porn.
What's scary is how close to home this hits. Out of stupid morbid curiosity, I almost downloaded some of this crap, just to see what it contained. In the end I said, "meh, a waste of bandwidth," turned up my nose at the unwashed masses and their silly jokes, and decided not to. Man, it turns out that it was a good thing I was a snob in those days! I could have ended up in the slammer!
Now it turns out that all those people, who I'd just thought of as "silly," were, according to the legal system of Australia, actually child pornographers. Yeesh.
I feel sorry for the poor bastard who just got lynched. The judge sounds like more of a pedophile than he is.
How can a boy though with one Y-chromosome become a girl? Does the chromosome change??
Sorry, your understanding of biology won't be tolerated here.
Whether you appear male or female is determined by hormone concentrations in the uterus. The Y chromosome only influences this indirectly, by somehow (the mechanism is not well understood) affecting the way these levels are set.
Then, how many people have genetic testing done to determine their chromosomal sex? If you were born with female genetalia, underwent puberty, and developed as a completely normal girl into a completely normal woman, why would you suspect that you had a Y chromosome?
It's easy to avoid kissing, etc. people who are actively having outbreaks. I avoid that too. But people with HSV are contagious at other times as well. The phrase is "asymptomatic viral shedding" -- and, actually, infected people are shedding more often than they are having outbreaks! So if you rule out people who have HSV-1, and not just those who are currently having an obvious outbreak, then you've ruled out 50% of the population. What do you do, ask a girl "Do you have any history of cold sores?" before you kiss her? It just doesn't seem practical to me.
The phrase "cold sore" is a misnomer that dates back to when people used to think that the sores had something to do with "getting a cold." And people keep using this phrase, I think, because it is a euphemism that lets them pretend that it's somehow "different" from that "other herpes."
Just do a Google image search for "cold sore." They tend to appear on the lips, outside the mouth. What you are describing sounds completely unrelated.
There's an issue with this of course. By the time you get older -- say, into your early 20s -- about 50% of the people your own age have oral HSV-1. What do you do, live in fear of contracting the virus? Don't kiss your date good night? Only consider romantic involvement with the 50% of the population that doesn't have HSV-1?
The problem is that the only way to avoid getting HSV-1 that isn't completely absurd is to just be lucky.
I believe that 90% of all adults are infected with the virus that causes Herpes. I know this is confusing, because of the confusion with genital herpes,
There are two types of herpes simplex virus: HSV-1 and HSV-2. Historically, HSV-1 has been called "oral herpes" and HSV-2 "genital herpes." But in fact, either can infect either location -- or other locations -- and both cause similar symptoms. (In industrialized nations, particularly among college students, most new cases of genital herpes are actually HSV-1. Ah, fellatio!)
Statistics: 50% of adults are seropositive for HSV-1. 25% of adults are seropositive for HSV-2.
It's erotica. Porn = explicit SEX, or naked people in a way designed to arouse you sexually. Erotica= just naked people, can be art.
Sorry, coder111, but that's not true. The word "erotic" does not mean "just happens to be naked;" it means sexually stimulating. It goes back to the word roots. From wikipedia:
Eros (Greek: á¼oeÏÏÏ), in Greek mythology, was the primordial god of lust, love, and intercourse
Now, the connotations of the word "erotic" are generally positive; it connotes not just sex, but sexual love. But this doesn't change the fact that, art portraying "naked people in a way designed to arouse you sexually" generally is erotica, by definition. (And the only reason I limit my statement with "generally" is that "erotic" implies connotations of love, which I suppose could be absent).
I agree. It is child porn. This cover is disgusting.
I have very mixed feelings about this. I was expecting a cut-and-dry case of unwarranted censorship as overcompensation by prudish government officials for their hidden pedophilic tendencies. But it's not that simple, and, as you note, the cover is not entirely innocent.
But I say that my feelings are "mixed" because going so far as to call it "child porn" somehow seems excessive. Society has decided that the production of child pornography is among the worst, most despicable crimes. Yet somehow this album cover does not seem to fit this characterization. It's borderline to be sure, but would it be appropriate to put, say, the band members or their photographer on a sex offender registry? I don't think so. Somehow, despite the fact that there is some sexualization of the girl in the image, the intent does not seem entirely pornographic.
As for the image itself:
On the one hand, the cover does have the girl in a pose that implies, through a degree or two of association, sex. It's not explicitly sexual; she's not engaging in or pantomiming any sex-act. But the image is also not of a girl who "just happens" to be nude. This brings me to my first question: What about the girl's pose evokes sex? Is it that the pelvis is tilted up, and is closer to the camera than the rest of the body? Is it that she is posed to provide an entirely unoccluded view of the (beginnings of) breasts, and to place them in the center of the image? I'm having a hard time saying what, in concrete terms, makes this pose sexualizing. Part of the reason I ask is that classical and Renaissance art adopted a number of stylized poses which, if we're honest, were partially erotic in intent. But this pose is slightly different. So I'm wondering what the important distinction is -- or if there is one.
More problematically, the broken glass effect radiates out from between the legs. This is obviously intended to emphasize the (occluded) vagina; I don't think there's any denying this.
So does all this add up to porn? I don't know. It's borderline. But then, that itself is part of the purpose of art; in fact, I sense that it may have become art's primary purpose: to explore boundaries, to shock, to offend sensibilities.
All of which makes it very hard for me to make a judgment here. But this, I suspect, is itself a reason not to censor: I think it's probably reasonable to say that censorship, if it is ever appropriate, is only so when there isn't any doubt; i.e., the benefit of the doubt should go to free speech.
were found with a bunch of issues of Playboy in their office
In the year 2008, soon to be 2009, is that really that scandalous?;-)
(Believe it or not, I actually once did cite Playboy in a literary analysis paper while at school: They carried an interview with Kurt Vonnegut and Joe Heller. No joke.)
(Franklin on the advantages of older women. He wrote a whole essay on the subject. His basic thesis was that the "juices" -- I think that was his word -- move down in the body as a woman ages...)
If you think the activities covered under the title "computer science" all involve a computer and working on it, well, you do not know what computer science is. If you think that "learning how computers really work" is the only possible motivation for doing computer science, well, you are wrong.
My favorite analogy: It's like calling astronomy "Telescope Science."
I disagree, because I don't think mean population density is a a useful number when talking about an area as large as the U.S. A snarky counterexample: Let 'A' be an area on the Earth's surface including Japan and 150 million square uninhabited kilometers of the Pacific Ocean. The mean population density in A is about 0.85 people/km^2. "Surely nobody should have broadband!" Clearly it's an absurd example, but you see my point.
For a more realistic example: New Jersey, for instance, has a population density of 454 people/km^2. Shouldn't really good broadband be rolling out there? (Perhaps this is a bad example: New Jersey does have FIOS in lots of places). Anyway, the point stands: The US has areas of high population density where it'd be perfectly feasible to roll out broadband on par with, e.g., South Korea, yet this seems to be happening slowly -- so there must be more variables than just population density. Perhaps other countries have figured out more effective regulatory structures than we have?
(I'll leave out the epsilon-delta discussion of the population density function that went on in my head right before I posted.)
Interesting!! Thanks for letting me know. Kubuntu Hardy -- which I'm using -- ships with a 2.6.24 kernel, but Intrepid comes with 2.6.27 -- so I'll get this when I get around to doing a distro upgrade. Nice!
"Better" might be a strong word, but I prefer rTorrent, since I can run it with screen and check up on it occasionally via ssh. Plus the system requirements are very low.
It's not without its negatives though. My biggest issue is that, because it uses mmap to write to files, it's incompatible with filesystems that do not support shared-writable mmaps (e.g., ntfs-3g, or anything else involving FUSE).
It's not that they remember fictional events, they simply report fictional events. Big difference. The child is aware that it's a fabrication.
What makes you say this? What I've read indicates just the opposite -- that children, and eyewitnesses in general, can honestly remember things that never happened. Memory is fallible, and the mind fills in gaps without you consciously realizing it.
I'd originally googled before posting, but didn't find the name; I thought I'd refer to my old textbooks next time I had access. But your post motivated me to try some new search strings, and this time I've found it, I think (or at least similar studies).
I believe the study I am referring to was carried out by Ceci (Cornell) and Bruck (McGill). [If not them, then perhaps it was a study by Goodman (UC Davis).] You might want to refer to this paper. In particular, see the section on p. 10 titled "Anatomically Detailed Dolls;" the study it describes is not exactly the same as I remember, but similar.
The kind of thing you describe is why any intellegent person should think twice about going into teaching. It's just too risky.
I've heard other people say this as well, and there's an element of truth to it. Unfortunately, what does this do to society? What happens when children -- especially boys -- have no male coaches, mentors, scout leaders... because it was "too risky?" This kind of paranoia is destroying communities.
Slashdot Mirror
That said, here's an algorithm I just made up that ten (metaphorical) bucks say will kick his GA's ass.
Undue bravado: easier said than done! I whipped up some code to do roughly what I described in my previous post, and though it works -- and perhaps more quickly than Alsling's GA, which took a million generations -- the results are much, much less impressive: It doesn't look as good, and it uses more polygons (however, I use triangles). I think a lot of it has to do with the fact that, although I can achieve a reasonable square error with my method, it neglects the low frequencies, which is actually what the eye pays attention to.
So: kudos, Roger Alsing.
Indeed.
First, a disclaimer, because I'm a nice guy: What this guy did is a neat sort of hack, and I approve. Many of us could have done this just as well, but this guy bothered and we didn't, so he deserves credit, and we should be happy for and with him. And it looks cool; that counts for a lot. So let us not rain on his parade. Sincere kudos are in order.
That said, here's an algorithm I just made up that ten (metaphorical) bucks say will kick his GA's ass. It stems from the fact that he's blending the polygons, which is, if it's additive blending he's using, a linear operation (if not, you can probably perform a transformation on the color values to turn it into something linear; e.g., take a log if it's multiplicative blending.)
Step 1: Generate a bajillion random images of polygons.
Step 2: Perform Graham-Schmidt orthonormalization on said images, while keeping track of the coefficients. In other words, do QR factorization.
Step 3: Project the Mona Lisa onto these vectors. I.e., premultiply it by Q^T.
Step 4: Throw out all of components of the projected Mona Lisa that fall below some threshold.
Step 5: Generate polygon-a-Lisa as a linear combination of images (from step 2) using the coefficients you computed in step 3 and kept in step 4.
The end! Isn't linear algebra fun?! :-) And of course you can rearrange the steps and do some similar variations of this to make the whole process less memory intensive; you don't need to actually store all these random images; you can instead do the projection and thresholding as you generate random polygons.
This has got to be the first time I've seen time-division multiplexing applied to sex.
What's frightening about this particular case is that I actually know about these files they're referring to -- and tons of people used to have them. Back when DirectConnect was a popular P2P app, I remember seeing files with names like "Simpsons porn" on lots of people's hard drives. I never thought much of it; obviously it was a crude joke and nothing more. It would never have even occurred to me to consider it child porn.
What's scary is how close to home this hits. Out of stupid morbid curiosity, I almost downloaded some of this crap, just to see what it contained. In the end I said, "meh, a waste of bandwidth," turned up my nose at the unwashed masses and their silly jokes, and decided not to. Man, it turns out that it was a good thing I was a snob in those days! I could have ended up in the slammer!
Now it turns out that all those people, who I'd just thought of as "silly," were, according to the legal system of Australia, actually child pornographers. Yeesh.
I feel sorry for the poor bastard who just got lynched. The judge sounds like more of a pedophile than he is.
How can a boy though with one Y-chromosome become a girl? Does the chromosome change?? Sorry, your understanding of biology won't be tolerated here.
Whether you appear male or female is determined by hormone concentrations in the uterus. The Y chromosome only influences this indirectly, by somehow (the mechanism is not well understood) affecting the way these levels are set.
Then, how many people have genetic testing done to determine their chromosomal sex? If you were born with female genetalia, underwent puberty, and developed as a completely normal girl into a completely normal woman, why would you suspect that you had a Y chromosome?
That's why I earned my EE degree. ;-)
The LM555 can be used in an astable multivibrator circuit!
It's easy to avoid kissing, etc. people who are actively having outbreaks. I avoid that too. But people with HSV are contagious at other times as well. The phrase is "asymptomatic viral shedding" -- and, actually, infected people are shedding more often than they are having outbreaks! So if you rule out people who have HSV-1, and not just those who are currently having an obvious outbreak, then you've ruled out 50% of the population. What do you do, ask a girl "Do you have any history of cold sores?" before you kiss her? It just doesn't seem practical to me.
Note to self: cold sore != frostbite
Indeed!
The phrase "cold sore" is a misnomer that dates back to when people used to think that the sores had something to do with "getting a cold." And people keep using this phrase, I think, because it is a euphemism that lets them pretend that it's somehow "different" from that "other herpes."
Just do a Google image search for "cold sore." They tend to appear on the lips, outside the mouth. What you are describing sounds completely unrelated.
There's an issue with this of course. By the time you get older -- say, into your early 20s -- about 50% of the people your own age have oral HSV-1. What do you do, live in fear of contracting the virus? Don't kiss your date good night? Only consider romantic involvement with the 50% of the population that doesn't have HSV-1?
The problem is that the only way to avoid getting HSV-1 that isn't completely absurd is to just be lucky.
Let me get this straight. If I always wear a condom I won't get Alzheimer's? :p
I think you'd have to wear a condom over your face, too.
You're mixing up 'yer virii. HSV vice HPV.
Yeah, the difference is in the statistics: While only the majority of people end up with HSV, damn near everybody ends up with HPV.
I believe that 90% of all adults are infected with the virus that causes Herpes. I know this is confusing, because of the confusion with genital herpes,
There are two types of herpes simplex virus: HSV-1 and HSV-2. Historically, HSV-1 has been called "oral herpes" and HSV-2 "genital herpes." But in fact, either can infect either location -- or other locations -- and both cause similar symptoms. (In industrialized nations, particularly among college students, most new cases of genital herpes are actually HSV-1. Ah, fellatio!)
Statistics: 50% of adults are seropositive for HSV-1. 25% of adults are seropositive for HSV-2.
It's erotica. Porn = explicit SEX, or naked people in a way designed to arouse you sexually. Erotica= just naked people, can be art.
Sorry, coder111, but that's not true. The word "erotic" does not mean "just happens to be naked;" it means sexually stimulating. It goes back to the word roots. From wikipedia:
Eros (Greek: á¼oeÏÏÏ), in Greek mythology, was the primordial god of lust, love, and intercourse
Now, the connotations of the word "erotic" are generally positive; it connotes not just sex, but sexual love. But this doesn't change the fact that, art portraying "naked people in a way designed to arouse you sexually" generally is erotica, by definition. (And the only reason I limit my statement with "generally" is that "erotic" implies connotations of love, which I suppose could be absent).
I agree. It is child porn. This cover is disgusting.
I have very mixed feelings about this. I was expecting a cut-and-dry case of unwarranted censorship as overcompensation by prudish government officials for their hidden pedophilic tendencies. But it's not that simple, and, as you note, the cover is not entirely innocent.
But I say that my feelings are "mixed" because going so far as to call it "child porn" somehow seems excessive. Society has decided that the production of child pornography is among the worst, most despicable crimes. Yet somehow this album cover does not seem to fit this characterization. It's borderline to be sure, but would it be appropriate to put, say, the band members or their photographer on a sex offender registry? I don't think so. Somehow, despite the fact that there is some sexualization of the girl in the image, the intent does not seem entirely pornographic.
As for the image itself:
On the one hand, the cover does have the girl in a pose that implies, through a degree or two of association, sex. It's not explicitly sexual; she's not engaging in or pantomiming any sex-act. But the image is also not of a girl who "just happens" to be nude. This brings me to my first question: What about the girl's pose evokes sex? Is it that the pelvis is tilted up, and is closer to the camera than the rest of the body? Is it that she is posed to provide an entirely unoccluded view of the (beginnings of) breasts, and to place them in the center of the image? I'm having a hard time saying what, in concrete terms, makes this pose sexualizing. Part of the reason I ask is that classical and Renaissance art adopted a number of stylized poses which, if we're honest, were partially erotic in intent. But this pose is slightly different. So I'm wondering what the important distinction is -- or if there is one.
More problematically, the broken glass effect radiates out from between the legs. This is obviously intended to emphasize the (occluded) vagina; I don't think there's any denying this.
So does all this add up to porn? I don't know. It's borderline. But then, that itself is part of the purpose of art; in fact, I sense that it may have become art's primary purpose: to explore boundaries, to shock, to offend sensibilities.
All of which makes it very hard for me to make a judgment here. But this, I suspect, is itself a reason not to censor: I think it's probably reasonable to say that censorship, if it is ever appropriate, is only so when there isn't any doubt; i.e., the benefit of the doubt should go to free speech.
were found with a bunch of issues of Playboy in their office
In the year 2008, soon to be 2009, is that really that scandalous? ;-)
(Believe it or not, I actually once did cite Playboy in a literary analysis paper while at school: They carried an interview with Kurt Vonnegut and Joe Heller. No joke.)
"All cats are gray in the dark."
(Franklin on the advantages of older women. He wrote a whole essay on the subject. His basic thesis was that the "juices" -- I think that was his word -- move down in the body as a woman ages...)
If you think the activities covered under the title "computer science" all involve a computer and working on it, well, you do not know what computer science is. If you think that "learning how computers really work" is the only possible motivation for doing computer science, well, you are wrong.
My favorite analogy: It's like calling astronomy "Telescope Science."
I disagree, because I don't think mean population density is a a useful number when talking about an area as large as the U.S. A snarky counterexample: Let 'A' be an area on the Earth's surface including Japan and 150 million square uninhabited kilometers of the Pacific Ocean. The mean population density in A is about 0.85 people/km^2. "Surely nobody should have broadband!" Clearly it's an absurd example, but you see my point.
For a more realistic example: New Jersey, for instance, has a population density of 454 people/km^2. Shouldn't really good broadband be rolling out there? (Perhaps this is a bad example: New Jersey does have FIOS in lots of places). Anyway, the point stands: The US has areas of high population density where it'd be perfectly feasible to roll out broadband on par with, e.g., South Korea, yet this seems to be happening slowly -- so there must be more variables than just population density. Perhaps other countries have figured out more effective regulatory structures than we have?
(I'll leave out the epsilon-delta discussion of the population density function that went on in my head right before I posted.)
they seek out information that validates their already-existing view of the world
Like Slashdot readers? (Obviously I'm guilty too.)
Interesting!! Thanks for letting me know. Kubuntu Hardy -- which I'm using -- ships with a 2.6.24 kernel, but Intrepid comes with 2.6.27 -- so I'll get this when I get around to doing a distro upgrade. Nice!
"Better" might be a strong word, but I prefer rTorrent, since I can run it with screen and check up on it occasionally via ssh. Plus the system requirements are very low.
It's not without its negatives though. My biggest issue is that, because it uses mmap to write to files, it's incompatible with filesystems that do not support shared-writable mmaps (e.g., ntfs-3g, or anything else involving FUSE).
I hadn't remembered the right keywords when I first posted; see my followup here.
It's not that they remember fictional events, they simply report fictional events. Big difference. The child is aware that it's a fabrication.
What makes you say this? What I've read indicates just the opposite -- that children, and eyewitnesses in general, can honestly remember things that never happened. Memory is fallible, and the mind fills in gaps without you consciously realizing it.
I'd originally googled before posting, but didn't find the name; I thought I'd refer to my old textbooks next time I had access. But your post motivated me to try some new search strings, and this time I've found it, I think (or at least similar studies).
I believe the study I am referring to was carried out by Ceci (Cornell) and Bruck (McGill). [If not them, then perhaps it was a study by Goodman (UC Davis).] You might want to refer to this paper. In particular, see the section on p. 10 titled "Anatomically Detailed Dolls;" the study it describes is not exactly the same as I remember, but similar.
The kind of thing you describe is why any intellegent person should think twice about going into teaching. It's just too risky.
I've heard other people say this as well, and there's an element of truth to it. Unfortunately, what does this do to society? What happens when children -- especially boys -- have no male coaches, mentors, scout leaders... because it was "too risky?" This kind of paranoia is destroying communities.