More than not exclusive - AAC is the audio implementation of the MPEG-4 codec, meaning that is an MPEG standard. The offerings from Apple (AAC and H.264) were impressive enough to warrant the basis of the next generation ISO/IEC MPEG standard group.
Mod parent up. The GP is obviously misplaced on Slashdot, given that he clearly doesn't know enough about electrical engineering to pass the first year at a decent university.
Knowing mathematically that something is possible and actually creating the first practical implementation are two entirely different things. Memresistors are a triumph of engineering.
My Macbook Pro is one of the most sturdy computers I have ever owned. White? No. However, it survived not one, not two, but 3 10 foot falls onto concrete. The case was beaten to hell. Looked as though id have to buy a new one - not so. After prying the twisted metal apart, I saw that every component in the thing was protected like a tank. There were individual aluminum housings and crumple zones all over. In fact, every part seemed fine.
So I took out a pair of pliers, a screwdriver, and putty knife and bent the thing back into what looks like its original form. Screwed it back together - still works perfectly after 3 years. I no longer question Macbook Pro build quality.
Woops, I meant to say that a monolithic kernel would run more slowly or take more memory if a feature is compiled in that you didn't use. The flexibility of the microkernel design allows you to simply remove the feature. Obviously, there are many instances where running a feature in kernel space will be much faster than running in user space.
"It should be fairly obvious why a "Hybrid Kernel" has a more modular design than a monolithic kernel."
Can't agree more. That said, I still tend to believe that the notion of running most of the core services in kernel space implies a monolithic kernel. I presented this as fact in earlier posts and apologize for that. It is my opinion that Windows NT's hybrid kernel has much more in common with a monolithic kernel and will be much more likely to take performance/memory hits when new system features are added, but I concede that others may have different feelings about hybrid kernels. I won't be so arrogant as to say that I'm right, as current experts are still debating how much a hybrid kernel actually represents the philosophy of a microkernel like GNU HURD.
What university is this? Mine (Northwestern) bases the entire CS systems curriculum around the GNU tool chain and unix/linux. We have a whole bunch of Red Hat machines for the department and then probably hundreds in the virtual lab.
I know you were probably just making a point, but a PhD does imply that you are an expert in your field. The only reason I'm pointing this out is because it seems in the cultural mindset to ignore PhDs with the philosophy, "Oh, they're just people the same as me. They don't know any more than I do and are just biased, sitting around in their ivory towers." I fear that this may be doing as much harm to science and academia as the cuts in funding.
Actually, no. When Tanenbaum was writing this, Windows NT was still in production, so he didn't have the full information about it. Windows NT is now commonly referred to as having a "hybrid kernel." In most of the cases that matter, it acts like a monolithic kernel structure. Wikipedia takes an even stronger stance:
Hybrid kernel is a kernel architecture based on combining aspects of microkernel and monolithic kernel architectures used in computer operating systems. The category is controversial due to the similarity to monolithic kernel; the term has been dismissed by some as just marketing.
I wouldn't go that far, but I would agree that, for this discussion, what matters is that most of the services reside in kernel space.
Considering that the kernel is highly relevant to how an operating system performs on benchmarks, I'd say both its design and structure are critical in determining speed. That said - you are correct. 7 is feature complete and we shouldn't see any big game changers very soon.
I think the problem here is that the ones who are saying that MMR causes autism are the exception, not the rule.
While I may have kind of agreed with you that many private practice physicians are behind on their science, I highly disagree that those in academia are. I worked in the Johns Hopkins medical institute and not a single one is devoid of scientific training.
The entire notion that those in academia are practicing substandard science is laughable - the science of medicine is in academia. You are talking about quacks who fear monger and are generally disregarded among the scientific community. They don't get a disproportionately loud voice in the media because they don't understand science - it's because the media loves sensationalist stories and conspiracy theories.
Or releasing countless, unending and contradictory studies about diet.
Ok, I see the problem. First, let me say that dietitians and nutritionists are not physicians - but really that's irrelevant. When you cite diet studies as an example of medicine, you are essentially playing the mainstream media game. You never hear about any of the scientific studies because the public wouldn't understand the science.
Academia is built on science, much the same way that the academic researchers in computer science are much more closely related with math and science than your average programmer. If someone asked you why computer science is really a "science," you wouldn't point them to the Cathedral and the Bazaar, would you? You would point them to P vs NP or Cellular Automata or any of the other million subsections of computer science theory. In the same way, just because a private practice physician may be behind on his biology, you should never assume the same of an academic researcher, or you will appear quite foolish at any reputable university.
That said, there are doctors with very good statistical skills. He is a molecular pathologist at Johns Hopkins University and one day he came up and asked me if I knew any open source statistical software.
I directed him to R http://www.r-project.org/, where he wrote and used an R implementation of the Kolmogorov-Smirnov test in his next paper. His BA is in chemistry, but it's essentially required that all physicians doing research (anything you ever see published in a journal) have a good knowledge of statistics and the math behind it.
Wow, is that you hate doctors or just that you can't respect any degree but a degree in your own field? Sure, I bet there are a few that meet your description, but it's pretty easy to know whether or not your doctor did "real" science. Look for someone who has done a http://en.wikipedia.org/wiki/Fellowship_(medicine) .
I will note that both of my parents are physicians (pathologists, if that matters) and know quite more about the science of biology than you would ever care to learn. Essentially, you're describing the class of doctors that are in it entirely for the money. If possible, try and find a physician who, at some point or another, worked in academia. My parents have said again and again that those who have worked in academia truly love medicine and science, and are taking a huge pay cut for it.
Copyright infringement is not only a civil case, it is also criminal. The GP is semantically wrong, but argumentally correct. Those who infringe copyright are breaking US federal law. http://www.copyright.gov/title17/92chap5.html#506
Personally, I believe that it's a good idea - they're just going about this the wrong way. Use an opt-in system. That way, those who wish to levy the p2p networks to get music can do so, while those who have nothing to do with music sharing can continue doing what they always have.
There are a number of universities that still teach it. I'm a CS major at Northwestern University - our intro CS class is Scheme and then most of the classes are either in C++ or C. Intro to Computer Systems (second year prereq for the degree) is taught with C, assembly, and even a little machine code.
That said, I still use Java for competition coding.
There is also Safari (on OS X) and Opera (on everything) in case you don't want your browser to run like a drugged quadruple amputee and leak memory like a sieve.
"Firstly, how much power does it use? "Three lightbulbs" says TFA, now as far as I'm aware the lightbulb is not a standard measurement for power consumption."
SWT is not maintained by Sun - it was originally created by IBM when they were first developing the Eclipse platform. It is now maintained and released by the Eclipse group. If you want 64-bit SWT compatibility, go yell at the Eclipse dev team.
Swing (as far as I can tell) runs perfectly fine on OS X 10.5.5 (64-bit). I will concede that, as a Java developer, the lack of an official 6.0 JDK did piss me off for a while.
Hmm i think we may just be using definitions. If you use a broadly defined hash function, yes, you could generate a "fuzzy" solution. This article and the previous article were both discussing the use of cryptographic hash functions (message digests), such as md5 and SHA. If you use message digests, there should be no "fuzzy" solution, but if you use a bad hash function it should be possible to generate a contrived solution for similar cases. While this is technically a hash function, I was referring to the general case of the "optimal" hashing function, which is defined as being injective (distinct input maps to distinct output). http://en.wikipedia.org/wiki/Injective_function
On the whole though, I think we're on the same page. Message digests cannot be practically expected to match anything but the exact image.
First, I just want to make clear that I never said hashing couldn't be useful - even hashing pictures. Looking for the exact picture is a huge time saver and very useful. My point was simply that hashing only works for exact matches.
That said, you seem to have another misconception that I tried to disabuse people of. There is no such thing as a "fuzzy" hash. We are using the word "hash" to mean the generally agreed one way cryptographic hashing function, also known as a message digest. In fact, the cryptographic design of message digest precludes anything like a "fuzzy" hash from existing. If you want to use an image comparison algorithm, that's a different story, but "fuzzy" hashes don't exist.
Bonus: AAC should properly be known as audio MPEG-4, meaning that it's both MPEG and ISO (http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=43345). AAC was designed to be the technical superior to mp3 (and it really is in potential) and is part of the next generation MPEG-4 standard.
Actually, no. This method does not work - which is what I said at the time. Because this misinformation is apparently still around, I decided to run a test.
I took a large file (1600x1200 px) and then applied a basic red-eye reducing algorithm to various spots on the image. The result: visually, exactly the same image.
Then I turned to my trusty Apple Preview. I resized each photo to 9% of its original size (144x108 pixels), and the proceeded to turn the color saturation down to 0 (black and white). I then saved each file in a compressionless TIFF format. Lastly, I computed the md5 hash for each file.
Let me say this once more, in case anyone else who blindly accepts anything they read on the internet has heard this: THIS TECHNIQUE DOES NOT WORK. To compare two SIMILAR images, one needs to use an image comparison algorithm - of which there are many. Hashing ONLY works on two images which are EXACTLY the same.
If you doubt the test or the results, I would be glad to email you all of my test pictures so you can see them and calculate their md5s for yourself.
Unless they can explicitly tie you as both knowledgeable of and complicit in said criminal acts - yup. Oh, and unless they could explicitly tie you to a "criminal enterprise" it wouldn't be RICO. Welcome to the United States justice system. Guilty beyond reasonable doubt or not guilty.
Light bulbs fail when they're turned on because there's a warmup spike as the electricity flows through the wire. This causes the tungsten to heat up more than it will when the circuit is complete, and break the circuit.
If your lightbulb fails it's because the filament has worn down - it's just that you usually find out about it when you turn the light bulb on or off.
Also, your analogy means nothing. Servers are nothing like light bulbs.
Hmm, no. Employees at an organization repeatedly violate federal laws because they're lazy and just want to get paid. The heads of ACORN don't really give a shit and don't pay attention to what's going on.
While possibly criminal (gross negligence), I wouldn't call this RICO or any kind of widespread liberal conspiracy (not to mention that McCain was also affiliated with ACORN previously). It's just a bunch of jerk-offs on the streets who fell behind in their quota but still wanted to get paid.
Slashdot Mirror
More than not exclusive - AAC is the audio implementation of the MPEG-4 codec, meaning that is an MPEG standard. The offerings from Apple (AAC and H.264) were impressive enough to warrant the basis of the next generation ISO/IEC MPEG standard group.
Check the official MPEG website if you don't believe me:
http://www.m4if.org/mpeg4/
Mod parent up. The GP is obviously misplaced on Slashdot, given that he clearly doesn't know enough about electrical engineering to pass the first year at a decent university.
Knowing mathematically that something is possible and actually creating the first practical implementation are two entirely different things. Memresistors are a triumph of engineering.
My Macbook Pro is one of the most sturdy computers I have ever owned. White? No. However, it survived not one, not two, but 3 10 foot falls onto concrete. The case was beaten to hell. Looked as though id have to buy a new one - not so. After prying the twisted metal apart, I saw that every component in the thing was protected like a tank. There were individual aluminum housings and crumple zones all over. In fact, every part seemed fine.
So I took out a pair of pliers, a screwdriver, and putty knife and bent the thing back into what looks like its original form. Screwed it back together - still works perfectly after 3 years. I no longer question Macbook Pro build quality.
Woops, I meant to say that a monolithic kernel would run more slowly or take more memory if a feature is compiled in that you didn't use. The flexibility of the microkernel design allows you to simply remove the feature. Obviously, there are many instances where running a feature in kernel space will be much faster than running in user space.
"It should be fairly obvious why a "Hybrid Kernel" has a more modular design than a monolithic kernel."
Can't agree more. That said, I still tend to believe that the notion of running most of the core services in kernel space implies a monolithic kernel. I presented this as fact in earlier posts and apologize for that. It is my opinion that Windows NT's hybrid kernel has much more in common with a monolithic kernel and will be much more likely to take performance/memory hits when new system features are added, but I concede that others may have different feelings about hybrid kernels. I won't be so arrogant as to say that I'm right, as current experts are still debating how much a hybrid kernel actually represents the philosophy of a microkernel like GNU HURD.
What university is this? Mine (Northwestern) bases the entire CS systems curriculum around the GNU tool chain and unix/linux. We have a whole bunch of Red Hat machines for the department and then probably hundreds in the virtual lab.
I know you were probably just making a point, but a PhD does imply that you are an expert in your field. The only reason I'm pointing this out is because it seems in the cultural mindset to ignore PhDs with the philosophy, "Oh, they're just people the same as me. They don't know any more than I do and are just biased, sitting around in their ivory towers." I fear that this may be doing as much harm to science and academia as the cuts in funding.
Actually, no. When Tanenbaum was writing this, Windows NT was still in production, so he didn't have the full information about it. Windows NT is now commonly referred to as having a "hybrid kernel." In most of the cases that matter, it acts like a monolithic kernel structure. Wikipedia takes an even stronger stance:
Hybrid kernel is a kernel architecture based on combining aspects of microkernel and monolithic kernel architectures used in computer operating systems. The category is controversial due to the similarity to monolithic kernel; the term has been dismissed by some as just marketing.
I wouldn't go that far, but I would agree that, for this discussion, what matters is that most of the services reside in kernel space.
The entire mechanism for building the OS is based on it being modular.
Of course it's not. Well, assuming you're talking about the Windows kernel anyway. http://oreilly.com/catalog/opensources/book/appa.html
Considering that the kernel is highly relevant to how an operating system performs on benchmarks, I'd say both its design and structure are critical in determining speed. That said - you are correct. 7 is feature complete and we shouldn't see any big game changers very soon.
I think the problem here is that the ones who are saying that MMR causes autism are the exception, not the rule.
While I may have kind of agreed with you that many private practice physicians are behind on their science, I highly disagree that those in academia are. I worked in the Johns Hopkins medical institute and not a single one is devoid of scientific training.
The entire notion that those in academia are practicing substandard science is laughable - the science of medicine is in academia. You are talking about quacks who fear monger and are generally disregarded among the scientific community. They don't get a disproportionately loud voice in the media because they don't understand science - it's because the media loves sensationalist stories and conspiracy theories.
Or releasing countless, unending and contradictory studies about diet.
Ok, I see the problem. First, let me say that dietitians and nutritionists are not physicians - but really that's irrelevant. When you cite diet studies as an example of medicine, you are essentially playing the mainstream media game. You never hear about any of the scientific studies because the public wouldn't understand the science.
Let me give you an example of an actual academic paper. I will not give my father's papers because I do not have his permission, however this is very similar to his area of expertise: http://www.sciencedirect.com/science?_ob=MImg&_imagekey=B8JDD-4RDPT4X-6-6&_cdi=43612&_user=1458830&_coverDate=02%2F29%2F2000&_sk=%23TOC%2343612%232000%23999339997%23677390%23FLA%23display%23Volume_66,_Issue_2,_Pages_i-ii,_347-753_(February_2000)%23tagged%23Volume%23first%3D66%23Issue%23first%3D2%23date%23(February_2000)%23&view=c&_gw=y&wchp=dGLbVzb-zSkzV&md5=acbce1898e1a993e7adeb4badc8ea004&ie=/sdarticle.pdf . I would like you to read this article and tell me where it lacks in science. In fact, here is the link to the New England Journal of Medicine. http://content.nejm.org/current.shtml My parents' house is full of the things. Look at the papers in "Original Articles." These are examples of science and I don't know how you can deny it.
Academia is built on science, much the same way that the academic researchers in computer science are much more closely related with math and science than your average programmer. If someone asked you why computer science is really a "science," you wouldn't point them to the Cathedral and the Bazaar, would you? You would point them to P vs NP or Cellular Automata or any of the other million subsections of computer science theory. In the same way, just because a private practice physician may be behind on his biology, you should never assume the same of an academic researcher, or you will appear quite foolish at any reputable university.
That said, there are doctors with very good statistical skills. He is a molecular pathologist at Johns Hopkins University and one day he came up and asked me if I knew any open source statistical software.
I directed him to R http://www.r-project.org/, where he wrote and used an R implementation of the Kolmogorov-Smirnov test in his next paper. His BA is in chemistry, but it's essentially required that all physicians doing research (anything you ever see published in a journal) have a good knowledge of statistics and the math behind it.
Wow, is that you hate doctors or just that you can't respect any degree but a degree in your own field? Sure, I bet there are a few that meet your description, but it's pretty easy to know whether or not your doctor did "real" science. Look for someone who has done a http://en.wikipedia.org/wiki/Fellowship_(medicine) .
I will note that both of my parents are physicians (pathologists, if that matters) and know quite more about the science of biology than you would ever care to learn. Essentially, you're describing the class of doctors that are in it entirely for the money. If possible, try and find a physician who, at some point or another, worked in academia. My parents have said again and again that those who have worked in academia truly love medicine and science, and are taking a huge pay cut for it.
Copyright infringement is not only a civil case, it is also criminal. The GP is semantically wrong, but argumentally correct. Those who infringe copyright are breaking US federal law. http://www.copyright.gov/title17/92chap5.html#506
Personally, I believe that it's a good idea - they're just going about this the wrong way. Use an opt-in system. That way, those who wish to levy the p2p networks to get music can do so, while those who have nothing to do with music sharing can continue doing what they always have.
There are a number of universities that still teach it. I'm a CS major at Northwestern University - our intro CS class is Scheme and then most of the classes are either in C++ or C. Intro to Computer Systems (second year prereq for the degree) is taught with C, assembly, and even a little machine code.
That said, I still use Java for competition coding.
There is also Safari (on OS X) and Opera (on everything) in case you don't want your browser to run like a drugged quadruple amputee and leak memory like a sieve.
Let's try some correct, instead of misleading, statistics: http://upload.wikimedia.org/wikipedia/commons/d/d7/US_annual_federal_deficits_over_receipts_1901_to_2006.svg
You need to read about what a budget surplus means.
"Firstly, how much power does it use? "Three lightbulbs" says TFA, now as far as I'm aware the lightbulb is not a standard measurement for power consumption."
That's about 8 Libraries of Congress.
SWT is not maintained by Sun - it was originally created by IBM when they were first developing the Eclipse platform. It is now maintained and released by the Eclipse group. If you want 64-bit SWT compatibility, go yell at the Eclipse dev team.
Swing (as far as I can tell) runs perfectly fine on OS X 10.5.5 (64-bit). I will concede that, as a Java developer, the lack of an official 6.0 JDK did piss me off for a while.
Informative? Let's try offtopic, mods.
Hmm i think we may just be using definitions. If you use a broadly defined hash function, yes, you could generate a "fuzzy" solution. This article and the previous article were both discussing the use of cryptographic hash functions (message digests), such as md5 and SHA. If you use message digests, there should be no "fuzzy" solution, but if you use a bad hash function it should be possible to generate a contrived solution for similar cases. While this is technically a hash function, I was referring to the general case of the "optimal" hashing function, which is defined as being injective (distinct input maps to distinct output). http://en.wikipedia.org/wiki/Injective_function
On the whole though, I think we're on the same page. Message digests cannot be practically expected to match anything but the exact image.
First, I just want to make clear that I never said hashing couldn't be useful - even hashing pictures. Looking for the exact picture is a huge time saver and very useful. My point was simply that hashing only works for exact matches.
That said, you seem to have another misconception that I tried to disabuse people of. There is no such thing as a "fuzzy" hash. We are using the word "hash" to mean the generally agreed one way cryptographic hashing function, also known as a message digest. In fact, the cryptographic design of message digest precludes anything like a "fuzzy" hash from existing. If you want to use an image comparison algorithm, that's a different story, but "fuzzy" hashes don't exist.
Bonus: AAC should properly be known as audio MPEG-4, meaning that it's both MPEG and ISO (http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=43345). AAC was designed to be the technical superior to mp3 (and it really is in potential) and is part of the next generation MPEG-4 standard.
Actually, no. This method does not work - which is what I said at the time. Because this misinformation is apparently still around, I decided to run a test.
I took a large file (1600x1200 px) and then applied a basic red-eye reducing algorithm to various spots on the image. The result: visually, exactly the same image.
Then I turned to my trusty Apple Preview. I resized each photo to 9% of its original size (144x108 pixels), and the proceeded to turn the color saturation down to 0 (black and white). I then saved each file in a compressionless TIFF format. Lastly, I computed the md5 hash for each file.
Result?
MD5 (smlimg3.tiff) = d300d23ce0ca2d6dcc7188665b1e2ada
MD5 (smlimg4.tiff) = a1cf7d59f9bf4ccceb6651c5f08750dd
Let me say this once more, in case anyone else who blindly accepts anything they read on the internet has heard this: THIS TECHNIQUE DOES NOT WORK. To compare two SIMILAR images, one needs to use an image comparison algorithm - of which there are many. Hashing ONLY works on two images which are EXACTLY the same.
If you doubt the test or the results, I would be glad to email you all of my test pictures so you can see them and calculate their md5s for yourself.
Unless they can explicitly tie you as both knowledgeable of and complicit in said criminal acts - yup. Oh, and unless they could explicitly tie you to a "criminal enterprise" it wouldn't be RICO. Welcome to the United States justice system. Guilty beyond reasonable doubt or not guilty.
Light bulbs fail when they're turned on because there's a warmup spike as the electricity flows through the wire. This causes the tungsten to heat up more than it will when the circuit is complete, and break the circuit.
If your lightbulb fails it's because the filament has worn down - it's just that you usually find out about it when you turn the light bulb on or off.
Also, your analogy means nothing. Servers are nothing like light bulbs.
Hmm, no. Employees at an organization repeatedly violate federal laws because they're lazy and just want to get paid. The heads of ACORN don't really give a shit and don't pay attention to what's going on.
While possibly criminal (gross negligence), I wouldn't call this RICO or any kind of widespread liberal conspiracy (not to mention that McCain was also affiliated with ACORN previously). It's just a bunch of jerk-offs on the streets who fell behind in their quota but still wanted to get paid.