If you're just doing network storage, there are many ways to skin a cat and they all work. WebDAV, like any of them, has some nice points (it's an open protocol, retrieval is a simple HTTP request) and some downsides (advanced versioning support still on the way, it's still in progress, setup on both client and server sides might be unfamiliar to many users). Go to www.webdav.org and learn more...
For the purposes of doing a centralized bookmark system (to steer us back on topic) the fact that retrieval is an HTTP request means that the much code is already in the browser if browser coders choose to write an extension. The upload part, IIRC, is along the lines of an extention to HTTP, so the code would be similar to code that's already in the browser. Since it runs over HTTP it'll be available just about everywhere that regular web access is, since most firewalls don't block standard web ports (yeah, I know, you can run any service on port 80).
This only matters if you're the person trying to implement the attack. From a perspective of securing the network all that matters is whether it's possible that a plugged-in USB device will autorun. Because if it's possible, even if it's hard for most people, it should be defended against.
There may be no thumbs/previews built in, but it would certainly be possible to write dynamic pages that scan the filesystem for images, generate thumbs with ImageMagick and put 'em on the page. Then you'd probably want to cache the thumbs to avoid wasting server cycles generating them, store a checksum of the original image in case it was changed, and as an extra feature add a description field to the database so anyone can add a description to non-descriptly named pictures via a simple web interface.
In principle, I certainly agree with you: a good way to solve this problem is to seperate the "file upload" problem from the "web interface" problem, because it's hard to make intuitive web-based file upload pages.
Trusted Computing is an application of encryption technology. If it's completely controlled by the owner of a computer system it can be a useful and powerful tool for security. It only is a problem for users if it's used against them by hardware and software makers, which is what many people fear will happen.
Re:more proof the RIAA/MPAA are insane
on
Death By DMCA
·
· Score: 1
So by your logic we are to say that the individual lobbyist trying to change a system of copyrights, patents and trademarks designed to encourage innovation into one designed to protect creators and deliverers of creative works is not to be criticized? He's only trying to feed his family? In his professional life I believe he is doing wrong to society. He could quit and find another job that doesn't erode my rights. I resent his work; it is his work and he's responsible for it.
The situation of military personnel is a bit different (but not completely different). They can't just quit because they don't agree with the mission, while most people can quit a job for any or no reason whatsoever. I would personally avoid taking a job where I had no free will or freedom to leave, but others might not. One might argue that military personnel carrying out an unjust war (by whatever standards one might determine a war to be unjust) are doing evil through their work. I wasn't alive during the time, but I think a lot of people expressed this type of feeling during the Vietnam war. The result was a lot of bitterness and a backlash against the demonization of individual troops that gave us the near-universal "support our troops" motto of today. Speaking hypothetically of a war that I felt to be unjust, I don't know if I would consider myself to "support" the troops in the same way that someone considering the war to be just would, even if I provided the same tangible types of support (in terms of donations and efforts to help families, etc.). If you really think about it it's a bit too complex an issue to slap on a bumper sticker.
I think most people would agree that the morality of both actions is conditional. Most people say that depriving someone of property is immoral. However, if someone is on a busy street going on a shooting rampage, depriving that person of his weapons (and perhaps even of his life) would be generally considered moral. Some really hardcore property rights-ers might say not even this is not moral and some communists might argue that depriving people of any concept of property is perfectly moral. There is a similar range of views on copyright: dada21 and his anarcho-capitalist posse think copying information is always OK and those that believe in the concept of "intellectual property" (the idea that ideas are property) might believe that the expiration of copyright terms is an act of violent force by the government.
There are some real differences between information and tangible objects, however. The big one is that information can be reproduced; in the case of digital representations of informtion it can even be exactly reproduced (copied). In most cases (with some exceptions) people that depend on physical goods depend on the fact that they have them. Copyright owners, on the other hand, generally exploit the fact that others cannot have copies of the works (with some exceptions, and I don't mean "exploit" as a loaded term). This idea of depending on others not having something as a basis for earning income to use to survive is a much newer one, and not quite as universally accepted. I also will state without giving much evidence that there are many more cases where copying information either (a) is encouraged by the information's creator or (b) does no harm whatsoever to the creator, even in the form of lost royalties/wages.
I don't think you can make the argument that copying information and depriving someone of property are equally moral, since the morality depends on many factors of the copying/deprivation that don't have analogues with respect to the other action. The only exception would be if you take a hard line and say that either action is always moral or immoral.
Damn, I always thought Times New Roman was kind of an ugly font (CMR, OTOH can look OK in PDFs if you use dvipdfm, but sure, it's not ideal for screen viewing). But that's just my own opinion. I was being a bit silly with the perfection thing; it has this aura of perfection around it because it's been the same powerful and flexible system for so many years, able to meet the demands of serious document creators. The one place I really think it shows its age, though, is interationalization (in this age of Unicode).
On the other hand, when they came up with Windows NT they implemented the Win32 API in the kernel to avoid the extra context switches required by a more userland-centric GUI (not to mention an X-style client-server model). They traded off flexibility and security (though, to be honest, security is hard to get when you require as much hardware access as GUIs do these days) for performance because they found the performance hit too much to bear, and at the time they just might have been right.
As some other posters have mentioned, I've seen Windows and Office run quickly on systems where, say, KDE under Linux and OpenOffice wouldn't be nearly that fast. On a P3 in the 400MHz range with plenty of RAM, Office feels just as "snappy" as GVIM.
In fact, some of the things you complain about in your post may have at one point been helped performance even though on modern hardware they hurt it. That doesn't make them things that shouldn't be complained about, but I hardly think that Microsoft is a company that can be said to not care about performance. They make their stuff perform very well in general on the computers they think it's important to run well on.
What's really desirable that runs on, say, GNU/Linux but not Windows? Well, there's the shell, some command-line utilities and window managers like FVWM and WindowMaker. If you're using those on Windows via Cygwin you'll probably realize that you're basically running two operating system userlands and library sets on top of one kernel and try to move towards the one you would rather work in (this is how I decided to switch to using the Free UNIXes almost exclusively). The desirable Windows programs that GNU/Linux users can only run under WINE are basically stand-alone apps (with the exception of web designers who need to run the vital Windows component that is Internet Explorer). Running, say, Google Earth under WINE doesn't make me pine for my Windows-using days, it's just another program sitting there on one of my virtual desktops.
Besides that fact, forget all the operating system evangelism. Manipulating users through limiting their choices is *so* closed-source. I went F/OSS because of software that gave me more choices and flexibility. Wine is such a piece of software.
The Computer Modern fonts (as used in TeX) are perfect! Or, at least, perfect enough that "These fonts are never going to change again".
I think the same thing is also true of TeX itself, but I may be wrong.:-P
Ah, WinME. Once in my younger days my brother was doing something at the computer and I was standing behind him. He had sent some files to the recycle bin, and right-clicked on the 'bin to purge them forever. He, as usual, immediately moved the mouse down and to the right and clicked, then clicked "OK" on the ensuing dialog box.
What he didn't notice was that the menu item he selected didn't say "Empty Recycle Bin", but rather "Delete Recycle Bin", and that the dialog was prompting him on whether he wanted the 'bin deleted permanently from the system. Strange quirk.
We never did figure out how to get the recycle bin back. That dialog wasn't joking, it was gone for good.
Now I'm not exactly sure why that was even an option that would ever come up. It only happened that one time... maybe we were pwn3d by a virus and we just didn't know it...
One of my favorite quotations about CS (though I'm not exactly a CS person) is that CS is as much about computers as astronomy is about telescopes (said by Dijkstra). To the degree that computers are like telescopes, you would have to be a student of programming, of computational techniques, to really be computer literate (as to be "telescope literate" you'd probably be studying astronomy at some level). But computers these days have uses other than the kind of computation that's really in the core realm of CS. Sure, those uses only exist because of enormous software systems that had to be programmed, but understanding those systems technically is fairly orthogonal to using them.
Computer literacy for most people doesn't involve using a computer like a telescope. It involves interacting with systems that someone's already programmed, not designing systems. Most people want to use their computers like a refridgerator: knowing that they need to put their food in there to keep it from spoiling, and that they should wrap some of it in airtight bags or something, but not knowing how the heat pump works. There are plenty of people with messy fridges, however, that know how heat pumps work. There are plenty of people that know a lot about computer architecture or programming that keep very disorganized PCs.
Maybe there are some elements of knowledge that are shared; an expert on fridges might better understand operating conditions, maintenance expectations, performance and limitations of a refridgerator (don't try to use the fridge to cool your house in the summer, for example). But that won't stop the expert from having rotten vegetables fermenting for months down in the bottom drawer. The same is true of computers. My technical knowledge of computers occasionally helps me use them, but more often it's just experience with various familiar computer interfaces: knowing the readline keybindings, all the wacky vim commands, and how to get through menus and websites in order to use GUIs or post to/.
runas is the same as su, it switches users. You must provide the password of the user you're switching to.
sudo is different. It allows users to do things they couldn't ordinarily do by merely authenticating as themselves. It allows for more fine-grained control over the things that each user can do than simply handing users the root password, and allows for better logging. That said, most people on their personal machines just have their sudo let them do anything.
But sudo is certainly more powerful than su/runas. I heard someone's trying to make a Windows version of sudo, but I don't know anything about it...
I used to set up all my systems so when you tried to "man $foo" it would print "Real men need no directions on $foo!" You had to type "woman $foo" to get the manpage. It got really annoying after a while. For some reason, people thought it was funny when I said, "woman gunzip". They were probably the same people that buy those "Girlz and Gunz" videos.
Speaking of that, are there any laptops out there with a true middle-click button? I'm not really in the mood to spend money on one right now, but trying to click both buttons at the same time to use X's "emulated" middle button is kind of a pain, and I'd keep it in mind for future laptop purchases.
"or break out of a virtual machine (is that possible through a module?)"
If you're in a virtual machine (at least a traditional virtual machine that doesn't require any special kernel support) you're not getting out without exploiting a security hole in the virtual machine program or processor. I don't think that Xen, user-mode Linux or any decent virtualization solution based on AMD or Intel's new processor extensions for virtualization would let you "break out" by merely changing the virtualized kernel either, but I might be wrong about that.
That said, the other reason you provided (obscuring the presence of malware) is something that's becoming increasingly common, though lots of obfuscation can be done by replacing userland system utilities and libraries. Combined with other techniques, disabling module support could help make unwanted programs easier to detect.
Sure, iTunes DRM is closed-source. But iTunes DRM doesn't cover the *gasp* analog hole. Or the digital hole, for that matter, in the operating system where a fake "audio driver" could be written to pipe output to a file. Windows is currently working on hardware-level DRM, with their "secure audio path" and "secure video path" stuff. This requires operating system support, and the *AA will soon realize that they want it and will be reluctant to license their music for use on any platform not supporting it.
I'm not exactly sure why a "secure audio/video path" would require a closed-source operating system; if the decryption is done on the device all the bits flowing through software are pretty meaningless anyway. But people have been talking about this stuff as if it's impossible for GNU/Linux and other Free systems to support it, so perhaps there must be closed parts of the kernel to implement it. I think that's what my grandparent post was talking about.
Err... the problem is that if the terrorists have won by banning all women they're pretty much going to lose by default some 60 years down the road or so.
I'm sure ALSA is great for something. Maybe high-end audio, sure. But for my mostly consumer-level needs I've had a lot more success with OSS. I am absolutely willing to admit that this is the fault of my own shoddy configuration: on my GNU/Linux system when using mplayer with ALSA the synchronization is always bad but using "OSS output" (which plays through ALSA anyway) works fine; in xmms ALSA output is garbled, and OSS output sounds great. All I can figure is that my.asoundrc must be bad but hell if I can find good resoureces on how to get good sound quality and sync with ALSA and still have dmix working. But for some reason the OSS emulation works perfectly. On my laptop with FreeBSD I've never once had to touch the OSS config (with the exception of telling it which kernel module to load at startup, then later just compiling it in statically) and it works perfectly with mixing.
Furthermore, is there a reason beyond just mindshare that there aren't drivers for these high-end cards (it seems that mindshare is the major reason that a lot of decent Linux/ALSA audio/music programs don't work with anything else, though the developers always put in their FAQs when asked about BSD support, "BSD doesn't support ALSA, which is superior," but don't actually back up this assertion. I've never seen a good defense of the "OSS sucks, ALSA rules" argument, so if you have a good reason that ALSA is technically better I'd like to hear it.
So now you've Microsoft Bob-ized the Web. If you want to transfer money to PayPal, you have to click on the ATM. Cute. But I think people have adapted to computers. They know, "I need to transfer money to my PayPal account", and they can pop open a new tab and do it. Adding an ATM just makes it take more time and effort, and forces them to make another new association because right now they don't associate ATMs with PayPal. PayPal doesn't actually give you cash.
Something like that only works if the users take the online world as some kind of immersive shopping environment. Call me crazy, but I don't think people want an immersive shopping environment any more than they wanted an immersive operating environment. They want their computers to give them the information they need to, in this case, make reasonable purchasing decisions. An immersive environment where they have to fight with and try to understand an additional metaphorical level on top of that whose only purpose in the end is to flash more ads in their face by forcing them to spend more time at the site. It won't work because it breaks what people like about the Internet: information that's easily, quickly, plainly available. It doesn't give you a shopping mall metaphor where you walk into each store and browse, it gives you a page listing more stores than you could remember the 3D coordinates for and how they charge for the specific item you want. It lets you specify what you're searching for in text rather than presenting a graphical environment with choices inherantly limited by the environment's creator. Yes, people REALLY DO LIKE TEXT over graphics when the choices are so vast, whether they know it or not.
Although some people consider these bugs to be fixed upstream, some Linux-centric projects use Linux-specific APIs. One of the most common of these APIs is ALSA. Though many projects support both OSS and ALSA, some important audio-related projects like Rosegarden only work with ALSA.
Other than that it's pretty much bugs, like you said. Some projects refuse to fix these bugs, calling themselves explicitly Linux-specific, basically saying, "if you want a BSD version maintain your own port" (which I can understand if it's an ALSA vs. OSS thing that would require lots of work to fix, but not if it was just simple portability stuff that would probably make it work more reliably on different Linux configurations anyway).
So if you have no DVD drive but a multi-GB flash drive on this thing you have to rip your DVDs in order to watch real movies on the device. That requires the use of "illegal" software (DVD rippers were at least at one point considered illegal, I don't know the exact details but they might not be illegal anymore; whether it's illegal or not it's certainly not something encouraged by the mainstream software vendors and content creators), and although banning software that might potentially be used for illegal things is stupid it's still a consideration. But more importantly it's a real pain in the ass. If I was going to go watch a movie on a portable device (not sure why I'd ever do this in the first place, but hypothetically...) I'd have to first rip it to my computer then copy the rip onto a flash drive. I'd need a flash drive with gigantic capacity and a few gigabytes to spare on my hard drive. If I want to watch very many movies I'll eventually have to buy more hard drives or burn data DVDs full of compressed rips and keep them in giant binders: I'll have turned into my brother!
I guess this is why we have UMD movies, although I can't imagine who'd ever want to buy a movie that can only be played on a PSP.
Any inaccuracies in this post are due to the fact that I very rarely buy or watch movies, especially on DVD (for me to buy a DVD, my appreciation of the quality of the work must overwhealm my distaste for their attempt to build DRM into the DVD format; future movie formats will only exaggerate my stance, by the looks of things, and I'll be confined to theater viewing, which honestly is fine with me because then I don't have to bother with all this expensive "home theater" crap).
The idea in quantum cryptography is that reading the photons would actually change them. So Bob would realize upon receipt of the photons that they'd been intercepted and they'd arrange to resend over some other channel.
How they'd communicate this arrangement is another issue.
Slashdot Mirror
If you're just doing network storage, there are many ways to skin a cat and they all work. WebDAV, like any of them, has some nice points (it's an open protocol, retrieval is a simple HTTP request) and some downsides (advanced versioning support still on the way, it's still in progress, setup on both client and server sides might be unfamiliar to many users). Go to www.webdav.org and learn more...
For the purposes of doing a centralized bookmark system (to steer us back on topic) the fact that retrieval is an HTTP request means that the much code is already in the browser if browser coders choose to write an extension. The upload part, IIRC, is along the lines of an extention to HTTP, so the code would be similar to code that's already in the browser. Since it runs over HTTP it'll be available just about everywhere that regular web access is, since most firewalls don't block standard web ports (yeah, I know, you can run any service on port 80).
This only matters if you're the person trying to implement the attack. From a perspective of securing the network all that matters is whether it's possible that a plugged-in USB device will autorun. Because if it's possible, even if it's hard for most people, it should be defended against.
There may be no thumbs/previews built in, but it would certainly be possible to write dynamic pages that scan the filesystem for images, generate thumbs with ImageMagick and put 'em on the page. Then you'd probably want to cache the thumbs to avoid wasting server cycles generating them, store a checksum of the original image in case it was changed, and as an extra feature add a description field to the database so anyone can add a description to non-descriptly named pictures via a simple web interface.
In principle, I certainly agree with you: a good way to solve this problem is to seperate the "file upload" problem from the "web interface" problem, because it's hard to make intuitive web-based file upload pages.
Trusted Computing is an application of encryption technology. If it's completely controlled by the owner of a computer system it can be a useful and powerful tool for security. It only is a problem for users if it's used against them by hardware and software makers, which is what many people fear will happen.
So by your logic we are to say that the individual lobbyist trying to change a system of copyrights, patents and trademarks designed to encourage innovation into one designed to protect creators and deliverers of creative works is not to be criticized? He's only trying to feed his family? In his professional life I believe he is doing wrong to society. He could quit and find another job that doesn't erode my rights. I resent his work; it is his work and he's responsible for it.
The situation of military personnel is a bit different (but not completely different). They can't just quit because they don't agree with the mission, while most people can quit a job for any or no reason whatsoever. I would personally avoid taking a job where I had no free will or freedom to leave, but others might not. One might argue that military personnel carrying out an unjust war (by whatever standards one might determine a war to be unjust) are doing evil through their work. I wasn't alive during the time, but I think a lot of people expressed this type of feeling during the Vietnam war. The result was a lot of bitterness and a backlash against the demonization of individual troops that gave us the near-universal "support our troops" motto of today. Speaking hypothetically of a war that I felt to be unjust, I don't know if I would consider myself to "support" the troops in the same way that someone considering the war to be just would, even if I provided the same tangible types of support (in terms of donations and efforts to help families, etc.). If you really think about it it's a bit too complex an issue to slap on a bumper sticker.
I think most people would agree that the morality of both actions is conditional. Most people say that depriving someone of property is immoral. However, if someone is on a busy street going on a shooting rampage, depriving that person of his weapons (and perhaps even of his life) would be generally considered moral. Some really hardcore property rights-ers might say not even this is not moral and some communists might argue that depriving people of any concept of property is perfectly moral. There is a similar range of views on copyright: dada21 and his anarcho-capitalist posse think copying information is always OK and those that believe in the concept of "intellectual property" (the idea that ideas are property) might believe that the expiration of copyright terms is an act of violent force by the government.
There are some real differences between information and tangible objects, however. The big one is that information can be reproduced; in the case of digital representations of informtion it can even be exactly reproduced (copied). In most cases (with some exceptions) people that depend on physical goods depend on the fact that they have them. Copyright owners, on the other hand, generally exploit the fact that others cannot have copies of the works (with some exceptions, and I don't mean "exploit" as a loaded term). This idea of depending on others not having something as a basis for earning income to use to survive is a much newer one, and not quite as universally accepted. I also will state without giving much evidence that there are many more cases where copying information either (a) is encouraged by the information's creator or (b) does no harm whatsoever to the creator, even in the form of lost royalties/wages.
I don't think you can make the argument that copying information and depriving someone of property are equally moral, since the morality depends on many factors of the copying/deprivation that don't have analogues with respect to the other action. The only exception would be if you take a hard line and say that either action is always moral or immoral.
Damn, I always thought Times New Roman was kind of an ugly font (CMR, OTOH can look OK in PDFs if you use dvipdfm, but sure, it's not ideal for screen viewing). But that's just my own opinion. I was being a bit silly with the perfection thing; it has this aura of perfection around it because it's been the same powerful and flexible system for so many years, able to meet the demands of serious document creators. The one place I really think it shows its age, though, is interationalization (in this age of Unicode).
On the other hand, when they came up with Windows NT they implemented the Win32 API in the kernel to avoid the extra context switches required by a more userland-centric GUI (not to mention an X-style client-server model). They traded off flexibility and security (though, to be honest, security is hard to get when you require as much hardware access as GUIs do these days) for performance because they found the performance hit too much to bear, and at the time they just might have been right.
As some other posters have mentioned, I've seen Windows and Office run quickly on systems where, say, KDE under Linux and OpenOffice wouldn't be nearly that fast. On a P3 in the 400MHz range with plenty of RAM, Office feels just as "snappy" as GVIM.
In fact, some of the things you complain about in your post may have at one point been helped performance even though on modern hardware they hurt it. That doesn't make them things that shouldn't be complained about, but I hardly think that Microsoft is a company that can be said to not care about performance. They make their stuff perform very well in general on the computers they think it's important to run well on.
What's really desirable that runs on, say, GNU/Linux but not Windows? Well, there's the shell, some command-line utilities and window managers like FVWM and WindowMaker. If you're using those on Windows via Cygwin you'll probably realize that you're basically running two operating system userlands and library sets on top of one kernel and try to move towards the one you would rather work in (this is how I decided to switch to using the Free UNIXes almost exclusively). The desirable Windows programs that GNU/Linux users can only run under WINE are basically stand-alone apps (with the exception of web designers who need to run the vital Windows component that is Internet Explorer). Running, say, Google Earth under WINE doesn't make me pine for my Windows-using days, it's just another program sitting there on one of my virtual desktops.
Besides that fact, forget all the operating system evangelism. Manipulating users through limiting their choices is *so* closed-source. I went F/OSS because of software that gave me more choices and flexibility. Wine is such a piece of software.
The Computer Modern fonts (as used in TeX) are perfect! Or, at least, perfect enough that "These fonts are never going to change again". I think the same thing is also true of TeX itself, but I may be wrong. :-P
Um, that's not perfect, that's grossly out of proportion -- and tacky to boot!
Ah, WinME. Once in my younger days my brother was doing something at the computer and I was standing behind him. He had sent some files to the recycle bin, and right-clicked on the 'bin to purge them forever. He, as usual, immediately moved the mouse down and to the right and clicked, then clicked "OK" on the ensuing dialog box.
What he didn't notice was that the menu item he selected didn't say "Empty Recycle Bin", but rather "Delete Recycle Bin", and that the dialog was prompting him on whether he wanted the 'bin deleted permanently from the system. Strange quirk.
We never did figure out how to get the recycle bin back. That dialog wasn't joking, it was gone for good.
Now I'm not exactly sure why that was even an option that would ever come up. It only happened that one time... maybe we were pwn3d by a virus and we just didn't know it...
One of my favorite quotations about CS (though I'm not exactly a CS person) is that CS is as much about computers as astronomy is about telescopes (said by Dijkstra). To the degree that computers are like telescopes, you would have to be a student of programming, of computational techniques, to really be computer literate (as to be "telescope literate" you'd probably be studying astronomy at some level). But computers these days have uses other than the kind of computation that's really in the core realm of CS. Sure, those uses only exist because of enormous software systems that had to be programmed, but understanding those systems technically is fairly orthogonal to using them.
/.
Computer literacy for most people doesn't involve using a computer like a telescope. It involves interacting with systems that someone's already programmed, not designing systems. Most people want to use their computers like a refridgerator: knowing that they need to put their food in there to keep it from spoiling, and that they should wrap some of it in airtight bags or something, but not knowing how the heat pump works. There are plenty of people with messy fridges, however, that know how heat pumps work. There are plenty of people that know a lot about computer architecture or programming that keep very disorganized PCs.
Maybe there are some elements of knowledge that are shared; an expert on fridges might better understand operating conditions, maintenance expectations, performance and limitations of a refridgerator (don't try to use the fridge to cool your house in the summer, for example). But that won't stop the expert from having rotten vegetables fermenting for months down in the bottom drawer. The same is true of computers. My technical knowledge of computers occasionally helps me use them, but more often it's just experience with various familiar computer interfaces: knowing the readline keybindings, all the wacky vim commands, and how to get through menus and websites in order to use GUIs or post to
runas is the same as su, it switches users. You must provide the password of the user you're switching to.
sudo is different. It allows users to do things they couldn't ordinarily do by merely authenticating as themselves. It allows for more fine-grained control over the things that each user can do than simply handing users the root password, and allows for better logging. That said, most people on their personal machines just have their sudo let them do anything.
But sudo is certainly more powerful than su/runas. I heard someone's trying to make a Windows version of sudo, but I don't know anything about it...
I used to set up all my systems so when you tried to "man $foo" it would print "Real men need no directions on $foo!" You had to type "woman $foo" to get the manpage. It got really annoying after a while. For some reason, people thought it was funny when I said, "woman gunzip". They were probably the same people that buy those "Girlz and Gunz" videos.
Speaking of that, are there any laptops out there with a true middle-click button? I'm not really in the mood to spend money on one right now, but trying to click both buttons at the same time to use X's "emulated" middle button is kind of a pain, and I'd keep it in mind for future laptop purchases.
"or break out of a virtual machine (is that possible through a module?)"
If you're in a virtual machine (at least a traditional virtual machine that doesn't require any special kernel support) you're not getting out without exploiting a security hole in the virtual machine program or processor. I don't think that Xen, user-mode Linux or any decent virtualization solution based on AMD or Intel's new processor extensions for virtualization would let you "break out" by merely changing the virtualized kernel either, but I might be wrong about that.
That said, the other reason you provided (obscuring the presence of malware) is something that's becoming increasingly common, though lots of obfuscation can be done by replacing userland system utilities and libraries. Combined with other techniques, disabling module support could help make unwanted programs easier to detect.
Sure, iTunes DRM is closed-source. But iTunes DRM doesn't cover the *gasp* analog hole. Or the digital hole, for that matter, in the operating system where a fake "audio driver" could be written to pipe output to a file. Windows is currently working on hardware-level DRM, with their "secure audio path" and "secure video path" stuff. This requires operating system support, and the *AA will soon realize that they want it and will be reluctant to license their music for use on any platform not supporting it.
I'm not exactly sure why a "secure audio/video path" would require a closed-source operating system; if the decryption is done on the device all the bits flowing through software are pretty meaningless anyway. But people have been talking about this stuff as if it's impossible for GNU/Linux and other Free systems to support it, so perhaps there must be closed parts of the kernel to implement it. I think that's what my grandparent post was talking about.
Err... the problem is that if the terrorists have won by banning all women they're pretty much going to lose by default some 60 years down the road or so.
Thou dost misquotest t3h 10rd!!!~1
The last word should be "b0x"!
And you say you know l33t arameic...
I'm sure ALSA is great for something. Maybe high-end audio, sure. But for my mostly consumer-level needs I've had a lot more success with OSS. I am absolutely willing to admit that this is the fault of my own shoddy configuration: on my GNU/Linux system when using mplayer with ALSA the synchronization is always bad but using "OSS output" (which plays through ALSA anyway) works fine; in xmms ALSA output is garbled, and OSS output sounds great. All I can figure is that my .asoundrc must be bad but hell if I can find good resoureces on how to get good sound quality and sync with ALSA and still have dmix working. But for some reason the OSS emulation works perfectly. On my laptop with FreeBSD I've never once had to touch the OSS config (with the exception of telling it which kernel module to load at startup, then later just compiling it in statically) and it works perfectly with mixing.
Furthermore, is there a reason beyond just mindshare that there aren't drivers for these high-end cards (it seems that mindshare is the major reason that a lot of decent Linux/ALSA audio/music programs don't work with anything else, though the developers always put in their FAQs when asked about BSD support, "BSD doesn't support ALSA, which is superior," but don't actually back up this assertion. I've never seen a good defense of the "OSS sucks, ALSA rules" argument, so if you have a good reason that ALSA is technically better I'd like to hear it.
So now you've Microsoft Bob-ized the Web. If you want to transfer money to PayPal, you have to click on the ATM. Cute. But I think people have adapted to computers. They know, "I need to transfer money to my PayPal account", and they can pop open a new tab and do it. Adding an ATM just makes it take more time and effort, and forces them to make another new association because right now they don't associate ATMs with PayPal. PayPal doesn't actually give you cash.
Something like that only works if the users take the online world as some kind of immersive shopping environment. Call me crazy, but I don't think people want an immersive shopping environment any more than they wanted an immersive operating environment. They want their computers to give them the information they need to, in this case, make reasonable purchasing decisions. An immersive environment where they have to fight with and try to understand an additional metaphorical level on top of that whose only purpose in the end is to flash more ads in their face by forcing them to spend more time at the site. It won't work because it breaks what people like about the Internet: information that's easily, quickly, plainly available. It doesn't give you a shopping mall metaphor where you walk into each store and browse, it gives you a page listing more stores than you could remember the 3D coordinates for and how they charge for the specific item you want. It lets you specify what you're searching for in text rather than presenting a graphical environment with choices inherantly limited by the environment's creator. Yes, people REALLY DO LIKE TEXT over graphics when the choices are so vast, whether they know it or not.
Although some people consider these bugs to be fixed upstream, some Linux-centric projects use Linux-specific APIs. One of the most common of these APIs is ALSA. Though many projects support both OSS and ALSA, some important audio-related projects like Rosegarden only work with ALSA.
Other than that it's pretty much bugs, like you said. Some projects refuse to fix these bugs, calling themselves explicitly Linux-specific, basically saying, "if you want a BSD version maintain your own port" (which I can understand if it's an ALSA vs. OSS thing that would require lots of work to fix, but not if it was just simple portability stuff that would probably make it work more reliably on different Linux configurations anyway).
So if you have no DVD drive but a multi-GB flash drive on this thing you have to rip your DVDs in order to watch real movies on the device. That requires the use of "illegal" software (DVD rippers were at least at one point considered illegal, I don't know the exact details but they might not be illegal anymore; whether it's illegal or not it's certainly not something encouraged by the mainstream software vendors and content creators), and although banning software that might potentially be used for illegal things is stupid it's still a consideration. But more importantly it's a real pain in the ass. If I was going to go watch a movie on a portable device (not sure why I'd ever do this in the first place, but hypothetically...) I'd have to first rip it to my computer then copy the rip onto a flash drive. I'd need a flash drive with gigantic capacity and a few gigabytes to spare on my hard drive. If I want to watch very many movies I'll eventually have to buy more hard drives or burn data DVDs full of compressed rips and keep them in giant binders: I'll have turned into my brother!
I guess this is why we have UMD movies, although I can't imagine who'd ever want to buy a movie that can only be played on a PSP.
Any inaccuracies in this post are due to the fact that I very rarely buy or watch movies, especially on DVD (for me to buy a DVD, my appreciation of the quality of the work must overwhealm my distaste for their attempt to build DRM into the DVD format; future movie formats will only exaggerate my stance, by the looks of things, and I'll be confined to theater viewing, which honestly is fine with me because then I don't have to bother with all this expensive "home theater" crap).
The idea in quantum cryptography is that reading the photons would actually change them. So Bob would realize upon receipt of the photons that they'd been intercepted and they'd arrange to resend over some other channel.
How they'd communicate this arrangement is another issue.