Pulse audio doesn't have drivers
on
Fedora 13 Is Out
·
· Score: 1
Can you please point me to this hardware compatibility list that you checked when you were "doing your homework". Because from what I can tell PulseAudio doesn't have any sound card drivers. It is just a sound server that provides network transparency and better mixing capabilites as an additional layer on top of the kernel sound support. It always uses an underlying layer like ALSA or OSS to talk to the hardware, as seen in this module diagram. Here is a full list of the PulseAudio modules - note that there are no direct hardware sinks, only sinks to other sound systems and piping capability.
Yeah, my impression being an on-and-off viewer was that they had a basic story line planned out initially, but then the network wanted to stretch the show out for longer, so they kept tacking things on until they had more loose ends then they could possibly resolve.
The chances of two given people having the same DNA fingerprint are tiny. So if the police already suspect someone of a crime, based on other reasons, then a DNA fingerprint match is good corroborating evidence.
However if you look for everyone that has the same DNA fingerprint as your sample in an entire city/state/country, you will almost certainly find multiple matches. In this case, the DNA match means absolutely nothing, but Jurors will treat it with the same weight as they did in the first case, because they don't understand statistics. Combine that with the fact that the defendant has had prior arrests (that's how he got in the database) and that is often enough to secure a conviction of an innocent man, even more so if he is poor and/or black.
Our justice system already convicts too many innocent people. Giving the government a tool that will result in more is a horrible idea.
Sorry, my wording was an over-statement. VP8 does allow a form of adaptive quantization, but it comes with overhead that is worse than both H.264 and Theora.
VP3 never had adaptive quanization, it was one of the few things added by Xiph before finalizing the Theora specification. (In general Xiph made all of the constant-value tables in VP3 configurable, including the quantization tables. While working on Vorbis they had found that this flexibility greatly improves the ability to do psy tuning, and wanted to have the same capability in Theora).
Thus Theora and H.264 are designed to allow you to specify different quantization for each block/macroblock. VP8 requires you to define regions for the different quantization blocks, and for complicated regions this adds more overhead then simply including a quantization table index would have.
Your link was a refutation to a rant written by Mans Rullgard. This was written by Jason Garrett-Glaser, and it is very objective. How about you base you judgment on the quality and validity of the assessment rather than guilt by association?
That video demonstrates the hallmark of On2 video encoders - they blur detail they can't encode. This is a nice trick that makes things look better at low bitrates where none of the codecs can encode all the detail, and would instead look blocky. However, this blurring tends to mask the fact that they aren't encoding as much detail as the other codecs. This becomes very obvious as the bitrates increase. On2 encoders remain blurry while others gain more detail.
That is why all of On2's promotional material uses low-bitrate examples. It is also why they hype their PSNR numbers - that is an image quality metric which does not always match human perception of image quality, and tends to favor blurred images.
Different VP8 encoders could be written that choose to optimize for detail rather than smoothness, but there is a limit to the amount of detail that can be preserved by the nature of the format. Jason's argument about the lack of adaptive quantization killing the ability of VP8 to match H.264 is pretty compelling to me.
Intra prediction is used to guess the content of a block without referring to other frames.
How the heck does that work? "Well I think Pinocchio's nose is growing in this frame, so I'll add some motion blur." - Pentium CPU. ???
It works the same way that standard image compression works. In real-life scenes, the adjacent pixels are highly correlated. As a simple example, a delta-predictor scans an image row-by-row and predicts that the next pixel will be the same as the previous, and then encodes the difference between the prediction and the pixel, rather than the pixel value itself.
While the total range of possible values increases (0 to 255 can now be -255 to +255), the entropy has decreased, so it can be encoded more efficiently using Huffman or Arithmetic encoding.
More importantly, the program that they are critcizing is the SM-3/Aegis program. It is a boost-phase interceptor deployed from ships off the coast of the adversary and whose goal is to hit the missile shortly after launch. There is a heck of a lot of ocean between North Korea and here where the deflected warhead could land.
The intercept rate for the Aegis program has been very impressive, and not really worth criticizing. His arguments about response time are valid, but we don't really have much option there.
Even discounting large inexpensive RAM, this feature still wouldn't require you to manually feed paper, like Colin_Smith suggested. The copy machine at my elementary school back in the early nineties had the ability to put a stack of papers on it, and it could make as many copies as you wanted purely mechanically. The only copier features that I have seen that needed a hard drive, are ones that no one uses because the interface is far too complicated for what should be a simple to use device.
Hey now, there are a large number of hardworking individuals in the government who are not elected and don't cast a vote. They have to work a lot harder for their bribes, and third party security information would make their lives much easier.
I can't think of any disadvantages to a faraday cage for a normal unrigged phone, other than the fact that you have to have one with you. The first cop on the scene probably won't, and will have to wait for the crime scene folks to show up. The phone could be remote wiped in that time. The disadvantages of removing the battery are that there may be info that is lost upon removing power and that not all phones have removable batteries.
A phone could be rigged to clear itself on boot and/or in the case of being without service for a certain period of time. Since you won't know which is the case then there really isn't much point in factoring it into your decision, though.
Some phones never truly turn off, and have the ability to be turned on remotely. The government was pushing for this feature, and now it has turned around and bit them. The only way to be certain that the black box you are carrying cannot communicate with the outside world is to remove the battery or stick it in a Faraday cage. Both methods have advantages and disadvantages.
And second, why not just fly current 737s a bit slower right now, to save on fuel?
It wouldn't help nearly as much. From another article I read, the main reason that they chose to fly slower wasn't to save additional fuel but because their aerodynamic design, while more efficient, puts higher strain on the engine and thus they need to fly slower to stay within safe margins.
you mention in your last sentence there "I don't see anything in copyright code specifically prohibiting the distribution of derivative works". That's not entirely true. Derivative works that are sufficiently original and transformative can actually qualify for their own copyright, and thus it would be illegal for the copyright holder of the original, or anyone else, to distribute it.
Yes if it does qualify for copyright protection then it obviously gets the same protection as any copyrighted work. However, if it is an unauthorized work, it does not qualify for copyright, and there seems to be a gap in the law concerning distribution of these. Their creation is clearly infringing, but there is nothing to say that their distribution is an additional offense. So by a strict reading of the law (which isn't always the correct reading) it appears that it is perfectly legal for third parties to distribute patches. I'd imagine they could probably be charged with some sort of contributory infringement like the filesharing software creators, though(or the DMCA for patches that disable DRM).
however, the bar for a derivative work is quite a bit higher than the bar for an original work to claim copyright.
Actually, while I was reading, I found a case decision that disagrees with this: Schrock v. Learning Curve International. Unfortunately I couldn't find a copy of the actual decision outside of LEXIS but here is some commentary.
as for the question of a derivative that does not contain any of the original work, i don't think there is any relevant case law since such a concept doesn't really make sense outside of the world of computers
I can think of a few examples that don't relate to computers. For example annotations are specifically mentioned as something that can be considered a derivative works, but if you were to publish annotations that simply referenced a line and page number, I'm not sure whether that would be a derivative work or not. Ie where is the line between annotation and commentary. Also, I have seen alternative soundtracks made for movies back in the VHS days. It is true that the computer world is rife with these situations ranging from the fully authorized (plugins) to grey area (game mods, vague library licenses) to fully unauthorized (haxies aka binary patches). It is a shame that there is no related precedent to look at.
After quantumplanet contradicted my post below, I spent some more time researching to see who was right. I'm still not completely sure what the legality of this case is, but it does look like authorization is an important factor.
Copyright of derivative works is granted according to 17 USC 103:
(a) The subject matter of copyright as specified by section 102 includes compilations and derivative works, but protection for a work employing preexisting material in which copyright subsists does not extend to any part of the work in which such material has been used unlawfully.
Since the preparation of derivative works is one of the exclusive rights granted to copyright holders, the preparation of a derivative work without authorization is unlawful. Therefore even if the crack has sufficient original content, it still wouldn't qualify for copyright protection under section 103.
So assuming that the patch is a derivative work, neither Myth nor anyone else (including Rockstar) would hold copyright on the patch. Therefore, Rockstar could legally distribute the cracked version as they hold copyright on the original and there is no protection for the patch.
In this case only way that Myth could claim copyright is if part of the patch was unrelated to the game (so it wouldn't be a derivative work), and original enough to qualify for it's own copyright. For example, logos sometimes qualify for copyright (in addition to trademark), and Myth's logo is included in the cracked version.
Where I am confused (and plan on researching some more) is surrounding derivative works that don't contain the original. I am having a hard time finding any relevant case law on whether or not a patch that does not include the original is a derivative work. I also don't know whether third parties could legally distribute the patch, derivative work or not. Since the patch doesn't contain the original work, and the patch itself is not protected by copyright, no distribution of copyright covered work it taking place. I don't see anything in copyright code specifically prohibiting the distribution of derivative works.
Rockstar most definitely does not own the copyright to the patch whether distributed alone or as a modified exe.
It is true that the pirates do not have the authority to create derivative works (except when it would be considered fair use), however, that does not mean that any infringing derivative works that are prepared are property of Rockstar. It just means that they are infringing, and that Rockstar can sue them. Furthermore, it does not mean that the patch is not protected under copyright law (assuming it is large enough to qualify for copyright). In the US all creative works that are eligible for copyright protection fall under copyright at the time they are created. In fact the authors of the patch probably could legitimately sue Rockstar for copyright infringement - they would just be counter-sued immediately.
This is the the same fallacy that people make when they say that linking code against GPL makes your code GPL. It doesn't. If you distribute the code, you can choose to make your code GPL to comply with the license or you can be sued for infringement. At no point can you be forced to make you code GPL, nor does the copyright of your code transfer to the author of the GPL code.
I would argue that non-CS majors should be taught a language that is most relevant to their field. Engineers: Matlab. Statistician: R Visual Design: Javascript Literature: Inform or TADS. Philosophers: Lisp
I don't like trackpads. They always register touches when I am typing and screw things up. Turning off tap-to-click helps somewhat, but they are still a pain. Oh and scrolling using the side of the pad never works for me. As much as I'd like the idea, the OS X multitouch trackpads are worse. They are constantly zooming when I meant to scroll, or scrolling when I meant to move the cursor or vice versa. I absolutely hate those things and they are another reason I will never buy an Apple laptop.
The eraser-caps are much nicer. They never move when you don't want them to. They can be very fast and precise once you get used to using them at high sensitivity. The middle button for scrolling is much nicer than any other implementation I have seen on a laptop.
Mice are much better than both for most things to be sure. But I don't like carrying around a mouse with me and the nubs are the best I have used on a laptop. Even on the desktop, if 99% of what you are doing is text, it is nice to be able to scroll and do quick cursor placements without taking your hands off the keyboard.
It is interesting how stereotypes differ in locations.
Where I grew up, we didn't have any quack chiropractors who thought they could heal cancer with back manipulations. It was just a synonym for back doctor. Physical therapists were who you went to to relearn how to walk after a serious accident. The idea of going to them for back pain or ergonomic advice is just bizarre to me.
Scientists are naturally curious people that want to understand how things work. From all accounts, ball lightning sounds like a plausible natural phenomenon, unlike other mysterious popularly reported things like ghosts, bigfoot, or aliens. Furthermore, if some reports are true, ball lightning has some very interesting properties, and understanding the physics behind it could have big implications.
Yeah, it is as easy to pirate as streaming radio, or images with "right-click protection". Which practically no one bothers to pirate. If you are going to pirate it is easier to let someone else do the reformatting for you and just get it off file sharing networks. I haven't checked but I wouldn't be surprised if many of these fonts are already available there.
WOFF is the answer to both questions. It is an open font format that allows browsers to download the font on demand, and all the browsers have committed to supporting it in their next release. It has no DRM, but since it isn't the same format as operating systems use, and the browser will be downloading it to a temporary directory behind the scenes, most users won't know that it is possible to copy the fonts - most don't even know how to install a TTF when you give it to them. The foundries have decided that being too restrictive about the use of fonts means that no one will use them, and have pretty much unanimously decided to support the WOFF format - which is what this article is about with all the tech info filtered out.
Slashdot Mirror
Can you please point me to this hardware compatibility list that you checked when you were "doing your homework". Because from what I can tell PulseAudio doesn't have any sound card drivers. It is just a sound server that provides network transparency and better mixing capabilites as an additional layer on top of the kernel sound support. It always uses an underlying layer like ALSA or OSS to talk to the hardware, as seen in this module diagram. Here is a full list of the PulseAudio modules - note that there are no direct hardware sinks, only sinks to other sound systems and piping capability.
Yeah, my impression being an on-and-off viewer was that they had a basic story line planned out initially, but then the network wanted to stretch the show out for longer, so they kept tacking things on until they had more loose ends then they could possibly resolve.
The chances of two given people having the same DNA fingerprint are tiny. So if the police already suspect someone of a crime, based on other reasons, then a DNA fingerprint match is good corroborating evidence.
However if you look for everyone that has the same DNA fingerprint as your sample in an entire city/state/country, you will almost certainly find multiple matches. In this case, the DNA match means absolutely nothing, but Jurors will treat it with the same weight as they did in the first case, because they don't understand statistics. Combine that with the fact that the defendant has had prior arrests (that's how he got in the database) and that is often enough to secure a conviction of an innocent man, even more so if he is poor and/or black.
Our justice system already convicts too many innocent people. Giving the government a tool that will result in more is a horrible idea.
Sorry, my wording was an over-statement. VP8 does allow a form of adaptive quantization, but it comes with overhead that is worse than both H.264 and Theora.
VP3 never had adaptive quanization, it was one of the few things added by Xiph before finalizing the Theora specification. (In general Xiph made all of the constant-value tables in VP3 configurable, including the quantization tables. While working on Vorbis they had found that this flexibility greatly improves the ability to do psy tuning, and wanted to have the same capability in Theora).
Thus Theora and H.264 are designed to allow you to specify different quantization for each block/macroblock. VP8 requires you to define regions for the different quantization blocks, and for complicated regions this adds more overhead then simply including a quantization table index would have.
Your link was a refutation to a rant written by Mans Rullgard. This was written by Jason Garrett-Glaser, and it is very objective. How about you base you judgment on the quality and validity of the assessment rather than guilt by association?
That video demonstrates the hallmark of On2 video encoders - they blur detail they can't encode. This is a nice trick that makes things look better at low bitrates where none of the codecs can encode all the detail, and would instead look blocky. However, this blurring tends to mask the fact that they aren't encoding as much detail as the other codecs. This becomes very obvious as the bitrates increase. On2 encoders remain blurry while others gain more detail.
That is why all of On2's promotional material uses low-bitrate examples. It is also why they hype their PSNR numbers - that is an image quality metric which does not always match human perception of image quality, and tends to favor blurred images.
Different VP8 encoders could be written that choose to optimize for detail rather than smoothness, but there is a limit to the amount of detail that can be preserved by the nature of the format. Jason's argument about the lack of adaptive quantization killing the ability of VP8 to match H.264 is pretty compelling to me.
Intra prediction is used to guess the content of a block without referring to other frames.
How the heck does that work? "Well I think Pinocchio's nose is growing in this frame, so I'll add some motion blur." - Pentium CPU. ???
It works the same way that standard image compression works. In real-life scenes, the adjacent pixels are highly correlated. As a simple example, a delta-predictor scans an image row-by-row and predicts that the next pixel will be the same as the previous, and then encodes the difference between the prediction and the pixel, rather than the pixel value itself.
While the total range of possible values increases (0 to 255 can now be -255 to +255), the entropy has decreased, so it can be encoded more efficiently using Huffman or Arithmetic encoding.
More importantly, the program that they are critcizing is the SM-3/Aegis program. It is a boost-phase interceptor deployed from ships off the coast of the adversary and whose goal is to hit the missile shortly after launch. There is a heck of a lot of ocean between North Korea and here where the deflected warhead could land.
The intercept rate for the Aegis program has been very impressive, and not really worth criticizing. His arguments about response time are valid, but we don't really have much option there.
Yes a shitty-ass computer with a horrible interface. I want to treat them like dumb input/output devices because that is what they are good at.
Even discounting large inexpensive RAM, this feature still wouldn't require you to manually feed paper, like Colin_Smith suggested. The copy machine at my elementary school back in the early nineties had the ability to put a stack of papers on it, and it could make as many copies as you wanted purely mechanically. The only copier features that I have seen that needed a hard drive, are ones that no one uses because the interface is far too complicated for what should be a simple to use device.
Hey now, there are a large number of hardworking individuals in the government who are not elected and don't cast a vote. They have to work a lot harder for their bribes, and third party security information would make their lives much easier.
I can't think of any disadvantages to a faraday cage for a normal unrigged phone, other than the fact that you have to have one with you. The first cop on the scene probably won't, and will have to wait for the crime scene folks to show up. The phone could be remote wiped in that time. The disadvantages of removing the battery are that there may be info that is lost upon removing power and that not all phones have removable batteries.
A phone could be rigged to clear itself on boot and/or in the case of being without service for a certain period of time. Since you won't know which is the case then there really isn't much point in factoring it into your decision, though.
Some phones never truly turn off, and have the ability to be turned on remotely. The government was pushing for this feature, and now it has turned around and bit them. The only way to be certain that the black box you are carrying cannot communicate with the outside world is to remove the battery or stick it in a Faraday cage. Both methods have advantages and disadvantages.
According to the article the proposed 737 replacement has standard wing length and is suitable for existing airports.
And second, why not just fly current 737s a bit slower right now, to save on fuel?
It wouldn't help nearly as much. From another article I read, the main reason that they chose to fly slower wasn't to save additional fuel but because their aerodynamic design, while more efficient, puts higher strain on the engine and thus they need to fly slower to stay within safe margins.
first off dude, it's quantumplacet, not planet.
sorry
you mention in your last sentence there "I don't see anything in copyright code specifically prohibiting the distribution of derivative works". That's not entirely true. Derivative works that are sufficiently original and transformative can actually qualify for their own copyright, and thus it would be illegal for the copyright holder of the original, or anyone else, to distribute it.
Yes if it does qualify for copyright protection then it obviously gets the same protection as any copyrighted work. However, if it is an unauthorized work, it does not qualify for copyright, and there seems to be a gap in the law concerning distribution of these. Their creation is clearly infringing, but there is nothing to say that their distribution is an additional offense. So by a strict reading of the law (which isn't always the correct reading) it appears that it is perfectly legal for third parties to distribute patches. I'd imagine they could probably be charged with some sort of contributory infringement like the filesharing software creators, though(or the DMCA for patches that disable DRM).
however, the bar for a derivative work is quite a bit higher than the bar for an original work to claim copyright.
Actually, while I was reading, I found a case decision that disagrees with this: Schrock v. Learning Curve International. Unfortunately I couldn't find a copy of the actual decision outside of LEXIS but here is some commentary.
as for the question of a derivative that does not contain any of the original work, i don't think there is any relevant case law since such a concept doesn't really make sense outside of the world of computers
I can think of a few examples that don't relate to computers. For example annotations are specifically mentioned as something that can be considered a derivative works, but if you were to publish annotations that simply referenced a line and page number, I'm not sure whether that would be a derivative work or not. Ie where is the line between annotation and commentary. Also, I have seen alternative soundtracks made for movies back in the VHS days. It is true that the computer world is rife with these situations ranging from the fully authorized (plugins) to grey area (game mods, vague library licenses) to fully unauthorized (haxies aka binary patches). It is a shame that there is no related precedent to look at.
After quantumplanet contradicted my post below, I spent some more time researching to see who was right. I'm still not completely sure what the legality of this case is, but it does look like authorization is an important factor.
Copyright of derivative works is granted according to 17 USC 103:
(a) The subject matter of copyright as specified by section 102 includes compilations and derivative works, but protection for a work employing preexisting material in which copyright subsists does not extend to any part of the work in which such material has been used unlawfully.
Since the preparation of derivative works is one of the exclusive rights granted to copyright holders, the preparation of a derivative work without authorization is unlawful. Therefore even if the crack has sufficient original content, it still wouldn't qualify for copyright protection under section 103.
So assuming that the patch is a derivative work, neither Myth nor anyone else (including Rockstar) would hold copyright on the patch. Therefore, Rockstar could legally distribute the cracked version as they hold copyright on the original and there is no protection for the patch.
In this case only way that Myth could claim copyright is if part of the patch was unrelated to the game (so it wouldn't be a derivative work), and original enough to qualify for it's own copyright. For example, logos sometimes qualify for copyright (in addition to trademark), and Myth's logo is included in the cracked version.
Where I am confused (and plan on researching some more) is surrounding derivative works that don't contain the original. I am having a hard time finding any relevant case law on whether or not a patch that does not include the original is a derivative work. I also don't know whether third parties could legally distribute the patch, derivative work or not. Since the patch doesn't contain the original work, and the patch itself is not protected by copyright, no distribution of copyright covered work it taking place. I don't see anything in copyright code specifically prohibiting the distribution of derivative works.
Rockstar most definitely does not own the copyright to the patch whether distributed alone or as a modified exe.
It is true that the pirates do not have the authority to create derivative works (except when it would be considered fair use), however, that does not mean that any infringing derivative works that are prepared are property of Rockstar. It just means that they are infringing, and that Rockstar can sue them. Furthermore, it does not mean that the patch is not protected under copyright law (assuming it is large enough to qualify for copyright). In the US all creative works that are eligible for copyright protection fall under copyright at the time they are created. In fact the authors of the patch probably could legitimately sue Rockstar for copyright infringement - they would just be counter-sued immediately.
This is the the same fallacy that people make when they say that linking code against GPL makes your code GPL. It doesn't. If you distribute the code, you can choose to make your code GPL to comply with the license or you can be sued for infringement. At no point can you be forced to make you code GPL, nor does the copyright of your code transfer to the author of the GPL code.
I would argue that non-CS majors should be taught a language that is most relevant to their field.
Engineers: Matlab.
Statistician: R
Visual Design: Javascript
Literature: Inform or TADS.
Philosophers: Lisp
I don't like trackpads. They always register touches when I am typing and screw things up. Turning off tap-to-click helps somewhat, but they are still a pain. Oh and scrolling using the side of the pad never works for me. As much as I'd like the idea, the OS X multitouch trackpads are worse. They are constantly zooming when I meant to scroll, or scrolling when I meant to move the cursor or vice versa. I absolutely hate those things and they are another reason I will never buy an Apple laptop.
The eraser-caps are much nicer. They never move when you don't want them to. They can be very fast and precise once you get used to using them at high sensitivity. The middle button for scrolling is much nicer than any other implementation I have seen on a laptop.
Mice are much better than both for most things to be sure. But I don't like carrying around a mouse with me and the nubs are the best I have used on a laptop. Even on the desktop, if 99% of what you are doing is text, it is nice to be able to scroll and do quick cursor placements without taking your hands off the keyboard.
It is interesting how stereotypes differ in locations.
Where I grew up, we didn't have any quack chiropractors who thought they could heal cancer with back manipulations. It was just a synonym for back doctor. Physical therapists were who you went to to relearn how to walk after a serious accident. The idea of going to them for back pain or ergonomic advice is just bizarre to me.
Scientists are naturally curious people that want to understand how things work. From all accounts, ball lightning sounds like a plausible natural phenomenon, unlike other mysterious popularly reported things like ghosts, bigfoot, or aliens. Furthermore, if some reports are true, ball lightning has some very interesting properties, and understanding the physics behind it could have big implications.
Can please tell me where in the AGPL this "implicit right" is derived. Because I read it and I think you're full of shit.
Yeah, it is as easy to pirate as streaming radio, or images with "right-click protection". Which practically no one bothers to pirate. If you are going to pirate it is easier to let someone else do the reformatting for you and just get it off file sharing networks. I haven't checked but I wouldn't be surprised if many of these fonts are already available there.
WOFF is the answer to both questions. It is an open font format that allows browsers to download the font on demand, and all the browsers have committed to supporting it in their next release. It has no DRM, but since it isn't the same format as operating systems use, and the browser will be downloading it to a temporary directory behind the scenes, most users won't know that it is possible to copy the fonts - most don't even know how to install a TTF when you give it to them. The foundries have decided that being too restrictive about the use of fonts means that no one will use them, and have pretty much unanimously decided to support the WOFF format - which is what this article is about with all the tech info filtered out.
This article has more info.