You can't possibly breed a "civilized" anything. A human baby today raised outside of civilization will not only fail to understand civilization, but will never be *able* to understand it once past a certain age. Certain parts of the brain don't develop in the necessary ways if they aren't stimulated early enough, like full language ability.
That all goes back to "nature vs nurture" arguments.
Sony using a customer's box for processing, at least without direct permission, would be theft. That box requires electricity to run, and if I'm not using the box, I damn well don't expect to be paying the electricity bill for it running at 100% capacity 24/7.
Despite the huge technical and social problems with this kind of change...
*If* it could happen, it would be great for many of us who want to block it out. Which is the purpose of the bill, of course.
Any mail that references an.xxx site can be blocked, browsers can be configured to refuse to load any resource from an.xxx site, search engines can refuse to search/list pages in.xxx domains, etc.
It's also possible for this to happen, I believe, to an extent; at the very least, due to the wonderful recently-showcased fact that the US controls the Internet naming infrastructure. Even foreign sites can be forced to comply by simply removing them from the top-level domains, and threatening to remove sites from top-level domains that host adult content.
One thing I'd worry about though is how one defines what is pornography and what isn't. Is a site that talks about STDs and safe-sex going to be labelled as adults-only by the religious right? Is a nudist colony site pornographic or simply counter-culture? Is a site that has "bad words" an adult site?
I would want to see a very clear, objective, strict, narrow definition of adult/pornographic content for this bill. i.e., "Images displaying sexual intercourse." (That is slightly too narrow, I'd think, but the intent should be clear.)
Hmm, something though to keep in mind is the existence of emulators and ports.
I never played the Final Fantasy games on the NES (despite being more than old enough to have done so, if I had wanted to), but I've played their ports on my GBA.
Likewise, I haven't owned a Super Nintendo in years, but I've been having a blast playing some of my childhood favorites on an SNES emulator on my brand-spanking-new state-of-the-art workstation with a USB gamepad.
The oldest of computer games are still playable in some form today. So long as we have computers in the future, and games aren't outlawed, I'm entirely confident that even the games we now consider archaic will still be playable long into the future.
The songs bought in iTMS are DRM protected. They cannot be played on computers or devices that are not authorized by iTMS. So, if iTMS ever goes under, or you somehow lose the ability to access iTMS (only machines that run iTunes can do this), your songs will then only play on the devices they've been authenticated on. And when those devices are obsolete, you're stuck with no way to get the songs to play on your new devices, unless they're Apple-approved.
Converting to any other format is going to cause a loss of quality. Even if you go to WAV or CD Audio, if you ever want to rip it back into some compressed format, you're going to lose quality.
Also, if you rip to WAV or CD, you lose all the meta-data for the track. So if you want to know the Artist, Title, and Album, you're going to have to re-enter that info on your own.
There's also no clean/easy way to export to MP3. Even if you jump through the hoops to do it, though, you're back to loss of quality.
I just went through the hell of exporting all my iTunes-purchased songs into Oggs so that I can play them on my Linux box, which has the nice sound system. That took quite a few burned CDs and I still haven't gotten the Oggs all retagged yet. Plus there's the quality issue, which while I've only noticed anything in a couple songs, that's still more quality issue than I would prefer.
AIGLX is simply support for accelerated indirect rendering.
Your X server already supports indirect rendering. The problem is that the OpenGL driver and library interfaces made it impossible for the X server to use the DRI driver. Only clients of the X server could use DRI.
IBM and others have been working on that issue for a while. The Red Hat work is in combination with those efforts, and has resulted in updates (some of them fairly complex) to the OpenGL driver system to allow the X server to access the OpenGL hardware directly. The X server needs to access the 3D hardware to make OpenGL-based compositing efficient.
Xglx does the exact same thing, but in a much more round-about way. Xglx is both an X server and an X client. Xglx simply uses the normal client-based DRI access. However, any clients attaching to Xglx lose DRI access. Instead they all do indirect rendering through Xglx, which then calls into DRI on behalf of the clients to get hardware acceleration.
Xglx also implements all of the core X drawing commands and RENDER using OpenGL, unlike X.org which uses the 2D drivers for those.
The new extension that the AIGLX page refers to is not an X extension. It's a GL extension. It is, in fact, the exact same GL extension that Xglx requires for decent performance.
AIGLX has quite a few advantages over Xglx, and not really any noteworthy disadvantages. It adds all of the capabilities that Xglx does, does it with less code changes, and retains the advantages and features offered by the existing 2D accelerated drivers. Any advantages you get from Xglx by having 2D rendering forced through the 3D driver can be done just as easily by client applications using 3D-enabled toolkits (Like GTK on Cairo on glitz).
Compiz requires a particular OpenGL extension that no hardware driver yet supports, so it is emulated in software by Xglx itself. NVIDIA helped define that extension, so it's safe to bet that future NVIDIA drivers will include the extension with full hardware acceleration. Likewise, the open source drivers are being updated to include the extension, though no official releases yet contain it. In a few years ATI might even include the extension in their official drivers, too.
In order to automatically compact the heap the entire process, including all of its libraries and other runtime components, must supply meta-data describing which values in memory are pointers and which are not. With C and C++, supplying that information is impossible. The closest you can get is the Boehm GC method of supplying memory layout information, but that doesn't apply to the stack. It also requires an enormous amount of effort without using some kind of preprocessor, and you are banned from using many common pointer tricks.
If you want a compacting collector, you're going to have to develop entirely in a language like Java, C#, Python, or another high level language.
The heap, where dynamic allocations occur, is only allowed to grow or to be truncated. An application cannot release memory in the middle of the heap without also releasing the memory at the end of the heap.
So let's say Firefox makes 10 one-page allocations, and frees the first 9. The memory layout might look something like: XXXXXXXXXU (X- unused, U- used)
Those 9 pages worth of memory aren't being used, but it's impossible to release them back to the OS.
Thankfully, there is some good news: when Firefox needs to allocate more memory, it can and will just reuse those 9 unused pages instead of allocating more memory from the OS and growing the heap.
The best solution to this problem is to use a compacting garbage collector. Which is something that Java and C# and other higher-level langauges can easily make use of (and many do use them), but which C and C++ can't really make use of given the complete lack of compiler support. That's one reason why a Java or C# app can actually out-perform a similar C/C++ app, especially with a good native-code compiler and an library implementation with a modern GC.
I have a nice speaker system setup with my Linux box. There's no way to play any of my $250 worth of iTMS-purchased songs on that machine. None. All of the files bought from iTMS are DRM-locked and can only be played on machines which are authorized through iTMS.
Unfortunately, there is no software for Linux which is capable of doing that. DVD Jon had released FairPlay and some other tools which could unlock those files, but Apple broke those utilities with the release of iTunes 6. You have to log into iTMS with a computer to generate a key for the computer. The tools can no longer log into iTMS, and so they cannot generate a key for my Linux box.
I can't copy the files off my Mac to my Linux box, I can't copy the files from my iPod to my Linux box, I can't copy the files to any machine that can't run iTunes, including any other portable music players made by companies other than Apple.
Sure, the several thousands songs I ripped from CDs to MP3s are fine and I can move those around, but at my current rate I would soon surpass the number of MP3s I have with the number of DRM-encumbered MP4s I have. (I don't plan on purchasing so much as one more song from iTMS until there is a way to transfer them to my other computers and devices.)
iTunes is fantastic if all you want to do is rip CDs onto your Mac or sync songs to an iPod. My iPod is breaking down (and is well out of warranty) and any replacement I buy will definitely not be an iPod, and my only Mac is an old iBook with horrendous sound ouput quality compared to the sound system on my Linux desktop. I've had to resort to burning my MP4s to CD (a lot of CDs), re-ripping them into Vorbis on the Linux box (losing some sound quality due to encoding the music twice), and then manually retagging all of the songs since the meta-data is lost when burned to CD. Whatever convenience I gained by using iTMS has now been lost.
iTunes *IS* DRM encumbered. Well, more accurately, iTMS is. [b]And that's what Songbird is competing with - the music store, not the music manager.[/b]
The beatings will continue until morale (and spelling) improves!
Re:And they don't even mention LARPing...
on
Masks in the Woods
·
· Score: 1
Indeed, I do LARPing as well. Our group owns a 40 acre plot of land, including buildings and trails. The role-playing experience when you are doing face-to-face character interaction is beyond ANYTHING you find in table-top or MMO scenarios.
Granted, if you're a role-player, be careful about just running out an joining a LARP. Most are just a bunch of dorky D&D nerds who though it'd be cool to dress up and hit each other with foam weapons. Finding a *good* LARP ain't easy.
Especially avoid NERO if you want role-playing. It's one of the largest medieval-ish LARPs around, and it almost completely lacks role-playing; people tend to sit around the campfire and talk about computer games or stuff like that. The combat at NERO is also absolutely disgustingly pathetic. The midwest has a number of good LARPs, though, and Europe has some truly massive and excellent LARPs. (One german LARP has some 5,000+ players on the field at a time.)
Given that the GPLv2 explitly states that any later version of the GPL may be applied instead of v2, *most* GPL projects will automatically be under the GPLv3.
The second processor is not identical to the first processor. It's also MMU-less. Stock Linux can't use the processor, and the various no-MMU versions of Linux have all been rejected by Linus for inclusion. (Note: My understanding of Linux MMU-less CPU support may be out of date.)
As much as I like Linux, I'm not sure that it's the best thing to put on this device. As someone highly interested in developing games for the system, I really don't care which OS is on it (or if there is an OS at all), so long as the SDK is complete and easy to use. From a gaming standpoint, I'd actually rather that there not be a full-on OS (when running games at least), I'd rather have the system boot straight into the game at ring0 and simply use a standpoint library for accessing the hardware features.
Granted, I also think the second CPU is kind of pointless anyway. The cost/space/power loss is pretty big for what the CPU could potentially give you; I'd much rather have seen those resources go into wireless support or a touch screen or a better GPU.
MD5 has not been invalidated for those uses. Checking the MD5 sum of an ISO download is not done for security purposes, it's done so that you can make sure you didn't get a bad byte or two somewhere in that 650MB. I mean, if hackers could upload a malware-filled ISO to the FTP server, they could upload a new MD5SUMS file too, right?
Something I find vastly amusing is that, using Wine or Cedega, it is generally easier to install Windows program on Linux than it is to install a Linux program on Linux.
Cedega: Pop in the CD, run the installer, run the updater (if its not automatic), done. Native: Open a terminal, run a shell script, watch it not quite work because your distro is 2 months newer than the software, manually hack the shell script to work, copy files over, manually create menu entries, download a tarball to update the game with, unpack the tarball, run the updater script, done. Native w/ Package: Find the package, realize you grabbed the wrong one because most people have no clue what the difference between.i386.rpm and.x86-64.rpm is, use a one-click install tool if you're lucky or open a terminal and manually install it if not, realize you are missing dependencies, install dependencies, done. Native w/ Package Search UI: Search through 10,000 poorly organized packages trying to find the right one (if you're lucky it is actually in the repository), install, done.
Most Open Source/Free Software/Linux folks seem to think that the last option is _clearly_ the best choice. I'm not so sure. Last I checked, NWN or Doom3 or Heretic II were not included in any RPM/DPKG repository, at least not any configured by default on any of the mainstream distributions.
The package selector interfaces in Synaptic or whatever is popular these days is also pretty much crap - when you have 10,000+ packages, you need something a little more efficient than a list with some hierarchial and practically meaningless categories like Amusements/Games.
Slashdot Mirror
You can't possibly breed a "civilized" anything. A human baby today raised outside of civilization will not only fail to understand civilization, but will never be *able* to understand it once past a certain age. Certain parts of the brain don't develop in the necessary ways if they aren't stimulated early enough, like full language ability.
That all goes back to "nature vs nurture" arguments.
You'd simply need a resistance-free wire. ... good luck with that.
Sony using a customer's box for processing, at least without direct permission, would be theft. That box requires electricity to run, and if I'm not using the box, I damn well don't expect to be paying the electricity bill for it running at 100% capacity 24/7.
Despite the huge technical and social problems with this kind of change...
.xxx site can be blocked, browsers can be configured to refuse to load any resource from an .xxx site, search engines can refuse to search/list pages in .xxx domains, etc.
*If* it could happen, it would be great for many of us who want to block it out. Which is the purpose of the bill, of course.
Any mail that references an
It's also possible for this to happen, I believe, to an extent; at the very least, due to the wonderful recently-showcased fact that the US controls the Internet naming infrastructure. Even foreign sites can be forced to comply by simply removing them from the top-level domains, and threatening to remove sites from top-level domains that host adult content.
One thing I'd worry about though is how one defines what is pornography and what isn't. Is a site that talks about STDs and safe-sex going to be labelled as adults-only by the religious right? Is a nudist colony site pornographic or simply counter-culture? Is a site that has "bad words" an adult site?
I would want to see a very clear, objective, strict, narrow definition of adult/pornographic content for this bill. i.e., "Images displaying sexual intercourse." (That is slightly too narrow, I'd think, but the intent should be clear.)
Hmm, something though to keep in mind is the existence of emulators and ports.
I never played the Final Fantasy games on the NES (despite being more than old enough to have done so, if I had wanted to), but I've played their ports on my GBA.
Likewise, I haven't owned a Super Nintendo in years, but I've been having a blast playing some of my childhood favorites on an SNES emulator on my brand-spanking-new state-of-the-art workstation with a USB gamepad.
The oldest of computer games are still playable in some form today. So long as we have computers in the future, and games aren't outlawed, I'm entirely confident that even the games we now consider archaic will still be playable long into the future.
The songs bought in iTMS are DRM protected. They cannot be played on computers or devices that are not authorized by iTMS. So, if iTMS ever goes under, or you somehow lose the ability to access iTMS (only machines that run iTunes can do this), your songs will then only play on the devices they've been authenticated on. And when those devices are obsolete, you're stuck with no way to get the songs to play on your new devices, unless they're Apple-approved.
Converting to any other format is going to cause a loss of quality. Even if you go to WAV or CD Audio, if you ever want to rip it back into some compressed format, you're going to lose quality.
Also, if you rip to WAV or CD, you lose all the meta-data for the track. So if you want to know the Artist, Title, and Album, you're going to have to re-enter that info on your own.
There's also no clean/easy way to export to MP3. Even if you jump through the hoops to do it, though, you're back to loss of quality.
I just went through the hell of exporting all my iTunes-purchased songs into Oggs so that I can play them on my Linux box, which has the nice sound system. That took quite a few burned CDs and I still haven't gotten the Oggs all retagged yet. Plus there's the quality issue, which while I've only noticed anything in a couple songs, that's still more quality issue than I would prefer.
AIGLX is simply support for accelerated indirect rendering.
Your X server already supports indirect rendering. The problem is that the OpenGL driver and library interfaces made it impossible for the X server to use the DRI driver. Only clients of the X server could use DRI.
IBM and others have been working on that issue for a while. The Red Hat work is in combination with those efforts, and has resulted in updates (some of them fairly complex) to the OpenGL driver system to allow the X server to access the OpenGL hardware directly. The X server needs to access the 3D hardware to make OpenGL-based compositing efficient.
Xglx does the exact same thing, but in a much more round-about way. Xglx is both an X server and an X client. Xglx simply uses the normal client-based DRI access. However, any clients attaching to Xglx lose DRI access. Instead they all do indirect rendering through Xglx, which then calls into DRI on behalf of the clients to get hardware acceleration.
Xglx also implements all of the core X drawing commands and RENDER using OpenGL, unlike X.org which uses the 2D drivers for those.
The new extension that the AIGLX page refers to is not an X extension. It's a GL extension. It is, in fact, the exact same GL extension that Xglx requires for decent performance.
AIGLX has quite a few advantages over Xglx, and not really any noteworthy disadvantages. It adds all of the capabilities that Xglx does, does it with less code changes, and retains the advantages and features offered by the existing 2D accelerated drivers. Any advantages you get from Xglx by having 2D rendering forced through the 3D driver can be done just as easily by client applications using 3D-enabled toolkits (Like GTK on Cairo on glitz).
Compiz requires a particular OpenGL extension that no hardware driver yet supports, so it is emulated in software by Xglx itself. NVIDIA helped define that extension, so it's safe to bet that future NVIDIA drivers will include the extension with full hardware acceleration. Likewise, the open source drivers are being updated to include the extension, though no official releases yet contain it. In a few years ATI might even include the extension in their official drivers, too.
Estimate: impossible.
In order to automatically compact the heap the entire process, including all of its libraries and other runtime components, must supply meta-data describing which values in memory are pointers and which are not. With C and C++, supplying that information is impossible. The closest you can get is the Boehm GC method of supplying memory layout information, but that doesn't apply to the stack. It also requires an enormous amount of effort without using some kind of preprocessor, and you are banned from using many common pointer tricks.
If you want a compacting collector, you're going to have to develop entirely in a language like Java, C#, Python, or another high level language.
The answer to that is pretty simple:
The heap, where dynamic allocations occur, is only allowed to grow or to be truncated. An application cannot release memory in the middle of the heap without also releasing the memory at the end of the heap.
So let's say Firefox makes 10 one-page allocations, and frees the first 9. The memory layout might look something like:
XXXXXXXXXU (X- unused, U- used)
Those 9 pages worth of memory aren't being used, but it's impossible to release them back to the OS.
Thankfully, there is some good news: when Firefox needs to allocate more memory, it can and will just reuse those 9 unused pages instead of allocating more memory from the OS and growing the heap.
The best solution to this problem is to use a compacting garbage collector. Which is something that Java and C# and other higher-level langauges can easily make use of (and many do use them), but which C and C++ can't really make use of given the complete lack of compiler support. That's one reason why a Java or C# app can actually out-perform a similar C/C++ app, especially with a good native-code compiler and an library implementation with a modern GC.
So what about the software that the user uses in order to ensure that only binaries from Bluehat are allowed to run?
I have a nice speaker system setup with my Linux box. There's no way to play any of my $250 worth of iTMS-purchased songs on that machine. None. All of the files bought from iTMS are DRM-locked and can only be played on machines which are authorized through iTMS.
Unfortunately, there is no software for Linux which is capable of doing that. DVD Jon had released FairPlay and some other tools which could unlock those files, but Apple broke those utilities with the release of iTunes 6. You have to log into iTMS with a computer to generate a key for the computer. The tools can no longer log into iTMS, and so they cannot generate a key for my Linux box.
I can't copy the files off my Mac to my Linux box, I can't copy the files from my iPod to my Linux box, I can't copy the files to any machine that can't run iTunes, including any other portable music players made by companies other than Apple.
Sure, the several thousands songs I ripped from CDs to MP3s are fine and I can move those around, but at my current rate I would soon surpass the number of MP3s I have with the number of DRM-encumbered MP4s I have. (I don't plan on purchasing so much as one more song from iTMS until there is a way to transfer them to my other computers and devices.)
iTunes is fantastic if all you want to do is rip CDs onto your Mac or sync songs to an iPod. My iPod is breaking down (and is well out of warranty) and any replacement I buy will definitely not be an iPod, and my only Mac is an old iBook with horrendous sound ouput quality compared to the sound system on my Linux desktop. I've had to resort to burning my MP4s to CD (a lot of CDs), re-ripping them into Vorbis on the Linux box (losing some sound quality due to encoding the music twice), and then manually retagging all of the songs since the meta-data is lost when burned to CD. Whatever convenience I gained by using iTMS has now been lost.
iTunes *IS* DRM encumbered. Well, more accurately, iTMS is. [b]And that's what Songbird is competing with - the music store, not the music manager.[/b]
The beatings will continue until morale (and spelling) improves!
Indeed, I do LARPing as well. Our group owns a 40 acre plot of land, including buildings and trails. The role-playing experience when you are doing face-to-face character interaction is beyond ANYTHING you find in table-top or MMO scenarios.
Granted, if you're a role-player, be careful about just running out an joining a LARP. Most are just a bunch of dorky D&D nerds who though it'd be cool to dress up and hit each other with foam weapons. Finding a *good* LARP ain't easy.
Especially avoid NERO if you want role-playing. It's one of the largest medieval-ish LARPs around, and it almost completely lacks role-playing; people tend to sit around the campfire and talk about computer games or stuff like that. The combat at NERO is also absolutely disgustingly pathetic. The midwest has a number of good LARPs, though, and Europe has some truly massive and excellent LARPs. (One german LARP has some 5,000+ players on the field at a time.)
Given that the GPLv2 explitly states that any later version of the GPL may be applied instead of v2, *most* GPL projects will automatically be under the GPLv3.
... and I put alcohol in my cereal I eat before going to work.
(And yes, I'm quite serious.)
The second processor is not identical to the first processor. It's also MMU-less. Stock Linux can't use the processor, and the various no-MMU versions of Linux have all been rejected by Linus for inclusion. (Note: My understanding of Linux MMU-less CPU support may be out of date.)
As much as I like Linux, I'm not sure that it's the best thing to put on this device. As someone highly interested in developing games for the system, I really don't care which OS is on it (or if there is an OS at all), so long as the SDK is complete and easy to use. From a gaming standpoint, I'd actually rather that there not be a full-on OS (when running games at least), I'd rather have the system boot straight into the game at ring0 and simply use a standpoint library for accessing the hardware features.
Granted, I also think the second CPU is kind of pointless anyway. The cost/space/power loss is pretty big for what the CPU could potentially give you; I'd much rather have seen those resources go into wireless support or a touch screen or a better GPU.
"but are you really interested in 90% of the crap they're turning out these days?"
:-/
No, definitely not, I'm interested in the other non-crappy 10%.
Which also do not run on Linux.
No, you're missing the point. If a song was *only* $0.99, that would imply that it was crap. All the popular songs will be more expensive, not less.
Do nothing.
MD5 has not been invalidated for those uses. Checking the MD5 sum of an ISO download is not done for security purposes, it's done so that you can make sure you didn't get a bad byte or two somewhere in that 650MB. I mean, if hackers could upload a malware-filled ISO to the FTP server, they could upload a new MD5SUMS file too, right?
Typical Open Source mindset...
"The release candidate is out... if you haven't upgraded yesterday, you're bloody obsolete!"
(I'm _joking_ people... put down the damn pitchforks already.)
"If I comment on this, it might damage your ego."
... it's whether or not you can roll the crit hits.
Possibly. But remember, it's not the size that counts...
(Just digging my well of depravity even deeper, aren't I?)
"And for all of the dirty D&D geeks out there, I already have a boyfriend."
Ah, but one must ask, are his D20s big enough for you?
(And there goes my "always respectful" track record.)
Something I find vastly amusing is that, using Wine or Cedega, it is generally easier to install Windows program on Linux than it is to install a Linux program on Linux.
.i386.rpm and .x86-64.rpm is, use a one-click install tool if you're lucky or open a terminal and manually install it if not, realize you are missing dependencies, install dependencies, done.
Cedega: Pop in the CD, run the installer, run the updater (if its not automatic), done.
Native: Open a terminal, run a shell script, watch it not quite work because your distro is 2 months newer than the software, manually hack the shell script to work, copy files over, manually create menu entries, download a tarball to update the game with, unpack the tarball, run the updater script, done.
Native w/ Package: Find the package, realize you grabbed the wrong one because most people have no clue what the difference between
Native w/ Package Search UI: Search through 10,000 poorly organized packages trying to find the right one (if you're lucky it is actually in the repository), install, done.
Most Open Source/Free Software/Linux folks seem to think that the last option is _clearly_ the best choice. I'm not so sure. Last I checked, NWN or Doom3 or Heretic II were not included in any RPM/DPKG repository, at least not any configured by default on any of the mainstream distributions.
The package selector interfaces in Synaptic or whatever is popular these days is also pretty much crap - when you have 10,000+ packages, you need something a little more efficient than a list with some hierarchial and practically meaningless categories like Amusements/Games.