Does anyone have any information as to what part of the law empowers SoundExchange to collect royalties for artists who do not have an explicit agreement with them? Personally, this is starting to seem like the most worrying thing about the state of US copyright, given how many ties I've heard cited between SoundExchange and the music distribution companies. It seems strange that this organization is allowed to collect on other people's work, especially since I've been given to understand that SoundExchange will not pay out royalties to the artist unless the artist in turn pays for a SoundExchange membership...
That's strange - I've been running a pure Bayesian filtering system (POPFile) for years now, and I haven't had to touch the training in ages... I'd say maybe one or two false negatives a month on a comcast.net account, and zero false positives.
Personally, I highly recommend POPFile, and hence probably other Bayesian filtering systems for e-mail processing... It's about 90% effective after a few days of training, and after additional training, we reach multiple-9's success rate VERY quickly.
It's not that Heinlein actually puts women down. In fact, I thoroughly agree with you that he essentially worships them. However, he does so in a fashion that some modern women find offensive, in that he assumes certain basic aspirations on biological grounds. As I recall (I haven't read Friday in a few years now), Friday is one of his worst books that way, largely because he takes the questionable step of narrating from the point of view of a female protagonist. The result is that the basic prejudices that he had (which, by the way, almost all members of either gender have in their mental conception of the opposite) come through in spades, and end up feeling almost directly sexist.
Where Heinlein differs from modern radical feminism is in his explicit upholding of the view that men and women have distinctly different roles to play in society. This break doesn't appear to be based in prejudice, but rather in his basic feeling that the average woman actually has far more significance, and thereby deserves better treatment, than the average man. Even this is not inviolable for him... several of his female characters break stereotypes right and left.
The primary way to defend Heinlein from these accusations, though, is to highlight how much POWER he attributes to women in each novel... Just as an on-the-fly interpretation, a one-sentence summary might be: "Men die for the world... Women live for it." Heinlein's world could almost survive without men; the essential role of women is beyond question, both biological and societal. The world revolves around women - men are an accessory of the real power. Hell, just look at his rather extreme views on sexuality and marriage... In each case, his societies give far more power to the women involved than the typical modern realization - and much more than did the society Heinlein was raised in! The Moon is a Harsh Mistress presents particularly good examples of this. In this context, and in every one of his non-dystopian visions of the future, one of the most despicable things any man can do is to force his views (and/or actions) on a woman.
Damn... I just don't know what we can do to fix this anymore. I'm honestly beginning to wonder if there's any chance of getting our freedom back. And the media coverage of all these problems? Nil.
How in the world do we get enough people to notice, at this point?
Also, are we college students really so apathetic now? The draft for the Vietnam War started riots, but there's next to no noise on campus over these problems - even at liberal schools...
I haven't lost hope yet, but how can we get the people of the United States to start caring again?
Not to turn this nasty... But you notice that the national debt as a percentage of GDP -peaked- during Clinton's term as president?
To be honest, judging the president on the basis of the economy is one of the silliest things I've ever heard of. I don't know of any direct power the President is able to exert over the economy, beyond appointing the people who matter - and generally, that just means finding someone smart and trusted (i.e. Greenspan) and then getting out of the way. I'm a liberal and generally aligned with the Democratic party, plus I very much liked Clinton, so I'll direct my cheap swipe that way... The economy was great under Clinton, wasn't it? Funny... every serious economist who supported trickle-down theory realized that its practice would take at least a decade to have any impact. Let's see... can we add and subtract to see who was President, and what was being done, at about a decade before that? Interesting, isn't it?
Again - the President has no direct power over the economy during his/her term in office. Indirect power, certainly, in declaring wars and establishing security, but these indirect methods result in only transient effects.
*ahem* You'll note the figures I gave aren't for the OLPC implementation, but the raw Wikipedia database dumps... I promise you, we're doing a bit better than that.
No pipe dream - unless I'm mistaken, the current OLPC ideal is to have a server in every school. However, the server won't act as a Wikipedia repository any more than each laptop will - except in that it may store additional Wikipedia "bundles" (groups of articles packaged as e-books) locally.
The figures I gave are for the entirety of the English Wikipedia.
Community server - I'm not sure what you mean, but since we can't always count on the laptops' having network connectivity, we're packaging a small (10-50 MB), static subset of Wikipedia directly on the laptops' storage, updateable on connection to the network (meaning Internet and/or local school server).
I presume it will be available once the selection/cleanup process is finished - after all, all software on the laptop is open-source.
The other offline versions of Wikipedia have a lot going for them, too - if you're interested, take a look at Wikipedia 1.0, and lend a hand if you care to. They're a very worthwhile effort, fully backed by Wikimedia and Jimmy Wales, and they're more specifically targeting CD- and DVD-sized versions.
Sorry, but I actually know something about this... My job this summer revolved around this issue, and seeing how much of Wikipedia we could cram into about 10 MB. (Hint: OLPC is using a subset of Wikipedia as its primary out-of-the-box reference material.)
The images on Wikipedia as of this January are about 76 GB in size. Now, assume we can switch to low-quality JPEGs and cut the size down to 5% of its current - about the size you'd get from switching all the images to black-and-white, in fact. Making that jump is a big assumption, but even that only gets you down to about 4 GB.
Text-wise, the Wikipedia database containing all current article info (no discussion pages, no history, etc.) is 1.7 GB - compressed. It's significantly larger when uncompressed.
This issue is being worked on. As I understand it, the closed wireless firmware is planned to be completely replaced in the next revision of the laptop.
Just a bit of info - as you say, the storage is a bit small. However, think of it this way - they abandoned hard drives because they fail too often. Think of the abuse these will undergo relative to most systems - sand, rough treatment, no air conditioning... These machines need to be hardy. So they went to flash storage, and as it is, the storage is probably one of the most expensive parts of the system. Expanding the storage available is definitely non-trivial in terms of cost.
As for the screen, the 1200x900 resolution is for black-and-white - this is important, as they're looking to maximize readability. In fact, they're looking for newspaper-level readability in bright light, which requires true revolutions in display technology - as I understand it, though, they've basically solved that problem. I've heard figures of about 690x520 as the effective color resolution, which seems perfectly reasonable.
Also, I've heard that cameras are getting awfully cheap (component-wise, anyway)... Since it doesn't need to do any of its own processing, I bet the camera (just the sensor) is less than $5-10 of the total cost. And think of the wow factor... remember, these are targeting kids! You get this kind of thing used by making it fun.
As I understand it, this is being very seriously considered... as I recall, they'd like to sell them at 3x cost in order to fully subsidize 2 laptops per laptop sold.
I think the simplest solution is to have the receiver be the only one who transmits his/her current measurements back. Given this, the sender can still stop sending at any point where the measurement disagrees, and the eavesdropper can't tell what measurements TO send back to prevent detection.
In the paper, though, it appears that the author essentially relies on simultaneity as the test to prevent the man-in-the-middle problem, or he's simply being vague on the topic. This seems extremely weak and unreliable... but I think the solution I proposed above might fix this.
On the topic of current-injection, though, I think I understand now... the only way to differentiate between which resistor is on which side is to inject on one wire and detect on the other, which results in a different current measurement on both sides, and the sign of the change is reversed on one side. This idea makes quite a bit of sense...
*Disclaimer* My expertise in the field of circuits is lacking - but I've studied it somewhat before.
There's something of a misunderstanding here - the author doesn't claim that the sides detect a change to find an eavesdropper, but that they find a *difference* in a measurement they make. This doesn't depend on the eavesdropper coming in in the middle.
Assuming that I'm understanding the paper correctly, the author claims that there's no way to distinguish which resistance is where when the resistors differ on the two sides without injecting some measureable current, which makes sense to me. One of Kirchoff's loop rules for circuitry basically says that charge is conserved in a closed circuit (current in = current out of any point) - and this is true to any currently measureable degree, as far as I know. In the normal setup, therefore, the sender and receiver should agree on the current in the circuit.
The next step is slightly shaky in my understanding - but I'm not sure if that's me or the concept. The paper claims that if the eavesdropper introduces a measureable current, it must create a measureable difference between the measured current values at the sender and receiver. Given this, if the sender and receiver consistently share the values they measure for the current at each step, over a public channel, they give away no information, but if there is an eavesdropper, they discover him/her immediately.
This seems to be a sensible idea, but I don't know enough to determine the truth of the step I mentioned above, so I can't tell if the proposed solution works.
Actually, it was JetBlue, and apparently they've done it from their founding - but the rest of the details are right. JetBlue refers to it as "homesourcing" - and Friedman holds it up as a perfect example of the positive domestic trends brought on by globalization.
Slashdot Mirror
Does anyone have any information as to what part of the law empowers SoundExchange to collect royalties for artists who do not have an explicit agreement with them? Personally, this is starting to seem like the most worrying thing about the state of US copyright, given how many ties I've heard cited between SoundExchange and the music distribution companies. It seems strange that this organization is allowed to collect on other people's work, especially since I've been given to understand that SoundExchange will not pay out royalties to the artist unless the artist in turn pays for a SoundExchange membership...
That's strange - I've been running a pure Bayesian filtering system (POPFile) for years now, and I haven't had to touch the training in ages... I'd say maybe one or two false negatives a month on a comcast.net account, and zero false positives. Personally, I highly recommend POPFile, and hence probably other Bayesian filtering systems for e-mail processing... It's about 90% effective after a few days of training, and after additional training, we reach multiple-9's success rate VERY quickly.
Disclaimer: I'm male.
It's not that Heinlein actually puts women down. In fact, I thoroughly agree with you that he essentially worships them. However, he does so in a fashion that some modern women find offensive, in that he assumes certain basic aspirations on biological grounds. As I recall (I haven't read Friday in a few years now), Friday is one of his worst books that way, largely because he takes the questionable step of narrating from the point of view of a female protagonist. The result is that the basic prejudices that he had (which, by the way, almost all members of either gender have in their mental conception of the opposite) come through in spades, and end up feeling almost directly sexist.
Where Heinlein differs from modern radical feminism is in his explicit upholding of the view that men and women have distinctly different roles to play in society. This break doesn't appear to be based in prejudice, but rather in his basic feeling that the average woman actually has far more significance, and thereby deserves better treatment, than the average man. Even this is not inviolable for him... several of his female characters break stereotypes right and left.
The primary way to defend Heinlein from these accusations, though, is to highlight how much POWER he attributes to women in each novel... Just as an on-the-fly interpretation, a one-sentence summary might be: "Men die for the world... Women live for it." Heinlein's world could almost survive without men; the essential role of women is beyond question, both biological and societal. The world revolves around women - men are an accessory of the real power. Hell, just look at his rather extreme views on sexuality and marriage... In each case, his societies give far more power to the women involved than the typical modern realization - and much more than did the society Heinlein was raised in! The Moon is a Harsh Mistress presents particularly good examples of this. In this context, and in every one of his non-dystopian visions of the future, one of the most despicable things any man can do is to force his views (and/or actions) on a woman.
Damn... I just don't know what we can do to fix this anymore. I'm honestly beginning to wonder if there's any chance of getting our freedom back. And the media coverage of all these problems? Nil. How in the world do we get enough people to notice, at this point? Also, are we college students really so apathetic now? The draft for the Vietnam War started riots, but there's next to no noise on campus over these problems - even at liberal schools... I haven't lost hope yet, but how can we get the people of the United States to start caring again?
Not to turn this nasty... But you notice that the national debt as a percentage of GDP -peaked- during Clinton's term as president? To be honest, judging the president on the basis of the economy is one of the silliest things I've ever heard of. I don't know of any direct power the President is able to exert over the economy, beyond appointing the people who matter - and generally, that just means finding someone smart and trusted (i.e. Greenspan) and then getting out of the way. I'm a liberal and generally aligned with the Democratic party, plus I very much liked Clinton, so I'll direct my cheap swipe that way... The economy was great under Clinton, wasn't it? Funny... every serious economist who supported trickle-down theory realized that its practice would take at least a decade to have any impact. Let's see... can we add and subtract to see who was President, and what was being done, at about a decade before that? Interesting, isn't it? Again - the President has no direct power over the economy during his/her term in office. Indirect power, certainly, in declaring wars and establishing security, but these indirect methods result in only transient effects.
*ahem* You'll note the figures I gave aren't for the OLPC implementation, but the raw Wikipedia database dumps... I promise you, we're doing a bit better than that.
No pipe dream - unless I'm mistaken, the current OLPC ideal is to have a server in every school. However, the server won't act as a Wikipedia repository any more than each laptop will - except in that it may store additional Wikipedia "bundles" (groups of articles packaged as e-books) locally.
The figures I gave are for the entirety of the English Wikipedia.
Community server - I'm not sure what you mean, but since we can't always count on the laptops' having network connectivity, we're packaging a small (10-50 MB), static subset of Wikipedia directly on the laptops' storage, updateable on connection to the network (meaning Internet and/or local school server).
I presume it will be available once the selection/cleanup process is finished - after all, all software on the laptop is open-source.
The other offline versions of Wikipedia have a lot going for them, too - if you're interested, take a look at Wikipedia 1.0, and lend a hand if you care to. They're a very worthwhile effort, fully backed by Wikimedia and Jimmy Wales, and they're more specifically targeting CD- and DVD-sized versions.
Sorry, but I actually know something about this... My job this summer revolved around this issue, and seeing how much of Wikipedia we could cram into about 10 MB. (Hint: OLPC is using a subset of Wikipedia as its primary out-of-the-box reference material.)
The images on Wikipedia as of this January are about 76 GB in size. Now, assume we can switch to low-quality JPEGs and cut the size down to 5% of its current - about the size you'd get from switching all the images to black-and-white, in fact. Making that jump is a big assumption, but even that only gets you down to about 4 GB.
Text-wise, the Wikipedia database containing all current article info (no discussion pages, no history, etc.) is 1.7 GB - compressed. It's significantly larger when uncompressed.
There - 6 GB total. And that's an achievement...
This issue is being worked on. As I understand it, the closed wireless firmware is planned to be completely replaced in the next revision of the laptop.
I'm actually pretty sure of this - it's not true of patents.
That's completely false... you're thinking of trademarks. There's no legal obligation to enforce a patent.
Just a bit of info - as you say, the storage is a bit small. However, think of it this way - they abandoned hard drives because they fail too often. Think of the abuse these will undergo relative to most systems - sand, rough treatment, no air conditioning... These machines need to be hardy. So they went to flash storage, and as it is, the storage is probably one of the most expensive parts of the system. Expanding the storage available is definitely non-trivial in terms of cost. As for the screen, the 1200x900 resolution is for black-and-white - this is important, as they're looking to maximize readability. In fact, they're looking for newspaper-level readability in bright light, which requires true revolutions in display technology - as I understand it, though, they've basically solved that problem. I've heard figures of about 690x520 as the effective color resolution, which seems perfectly reasonable. Also, I've heard that cameras are getting awfully cheap (component-wise, anyway)... Since it doesn't need to do any of its own processing, I bet the camera (just the sensor) is less than $5-10 of the total cost. And think of the wow factor... remember, these are targeting kids! You get this kind of thing used by making it fun.
As I understand it, this is being very seriously considered... as I recall, they'd like to sell them at 3x cost in order to fully subsidize 2 laptops per laptop sold.
I think the simplest solution is to have the receiver be the only one who transmits his/her current measurements back. Given this, the sender can still stop sending at any point where the measurement disagrees, and the eavesdropper can't tell what measurements TO send back to prevent detection.
In the paper, though, it appears that the author essentially relies on simultaneity as the test to prevent the man-in-the-middle problem, or he's simply being vague on the topic. This seems extremely weak and unreliable... but I think the solution I proposed above might fix this.
On the topic of current-injection, though, I think I understand now... the only way to differentiate between which resistor is on which side is to inject on one wire and detect on the other, which results in a different current measurement on both sides, and the sign of the change is reversed on one side. This idea makes quite a bit of sense...
*Disclaimer* My expertise in the field of circuits is lacking - but I've studied it somewhat before.
There's something of a misunderstanding here - the author doesn't claim that the sides detect a change to find an eavesdropper, but that they find a *difference* in a measurement they make. This doesn't depend on the eavesdropper coming in in the middle.
Assuming that I'm understanding the paper correctly, the author claims that there's no way to distinguish which resistance is where when the resistors differ on the two sides without injecting some measureable current, which makes sense to me.
One of Kirchoff's loop rules for circuitry basically says that charge is conserved in a closed circuit (current in = current out of any point) - and this is true to any currently measureable degree, as far as I know. In the normal setup, therefore, the sender and receiver should agree on the current in the circuit.
The next step is slightly shaky in my understanding - but I'm not sure if that's me or the concept. The paper claims that if the eavesdropper introduces a measureable current, it must create a measureable difference between the measured current values at the sender and receiver.
Given this, if the sender and receiver consistently share the values they measure for the current at each step, over a public channel, they give away no information, but if there is an eavesdropper, they discover him/her immediately.
This seems to be a sensible idea, but I don't know enough to determine the truth of the step I mentioned above, so I can't tell if the proposed solution works.
I suspect Google can map out NYC public transit fairly easily, actually... HopStop seems to do it quite well! Link: http://www.hopstop.com/
Actually, it was JetBlue, and apparently they've done it from their founding - but the rest of the details are right. JetBlue refers to it as "homesourcing" - and Friedman holds it up as a perfect example of the positive domestic trends brought on by globalization.