I wonder why this was modded "troll"? "Redundant" I could see, but otherwise everything in it is quite true and well-reasoned. Was it because of your user name and sig? Or was it because you expressed a preference for Windows over Linux? (I found that hillarious, BTW, and completely agree with the sentiment. Of course, to each his own and all that.)
I am reading the patent, and I really don't understand how this could even be interpreted as being a patent on the concept of XML. It is a patent on a system that uses XML, as is pretty explicitly stated in the abstract.
Whether the patent itself is overly broad is up for debate. However, you can't just quote one line from the abstract and claim that the patent applies to everything in the universe that fits that one line. There is a reason for the body of the patent: to describe the specifics of the invention they are patenting.
Google blocked the "litigious bastards"->sco.com googlebomb. Why? Probably because, even though they hate SCO, they'd like to do whatever they can to avoid a legal fight.
Most googlebombs work within 48 hours. Several days after this one was suggested, it still wasn't working. At that point, I (yes, I myself) went around to a bunch of sites and suggested they change the link to caldera.com. Most of them ignored me, but I guess enough switched to make it work. Yay!
No no, you misread. He didn't say 50%. He said.50%. That comes to $312.50 a year. That's probably a lot more than what the guy makes in his parents' basement, though.
As a server, Linux is more stable. As a desktop, Windows (2k or XP, not 9x) is more stable. This is why I use Linux on my servers and Windows XP on my desktop. (However, I intend to switch to FreeBSD on the servers and Mac OSX on the desktop when I get the chance.)
Bull. The fact that XP froze on you proves only that you are not running a stable platform. Try ASUS A7N8X mobo + NVidia reference video with adequate cooling and you'll never lock up, period. At least, that's what I have, and XP never locks on me. And no, it's not that I don't stress it. I run games galore, develop software, run servers, use two monitors, have two RAID arrays installed (a RAID-5 and a RAID-0), etc. etc.
(And if Linux is perfectly stable for you, all that proves is that Linux happens to have better drivers for your particular hardware, or perhaps drivers that use fewer hardware features. In any case, that's the hardware manufacturer's fault, not XP's.)
Bad apps do not crash XP. Bad hardware and drivers crash XP. You must be thinking of 9x.
First, you'd have to save it to your hard drive, clicking on it wouldn't work
Most Windows e-mail clients will not open an executable when you simply click on it. In fact, they usually open multiple warning windows saying, essentially, "If you run this, you are a complete and total moron. Are you a moron? [YES] [NO]".
Then you'd need to "chmod +x" it
This provides about the same amount of protection as said warning windows. In order to run the program, you have to be fully aware that you are trying to run an executable. Having to chmod +x it is just an inconvenience, really.
and then you could run it as your user, in which case it can infect only things associated with that user.
Seriously, how many people read their e-mail on multi-user machines? Yes, I know there are some, but it's rare. In most cases, the person reading the e-mail is the only user of the particular machine they are on, and so having their own account totally trashed isn't really any better than just having the whole computer trashed.
Besides that, most viruses these days can accomplish all of their goals just as easily from a user account as they can from root. Typically, this involves propogating itself (requires only network access) and then carrying out some form of DDoS (again, requires only network access). Who needs root?
(Of course, on Windows, if you're smart, you're probably running ZoneAlarm, which will tell you when a program tries to access the internet and allow you to deny it that access. I am not familiar with any similar software for Linux. Though, if you're smart, you aren't running attachments anyway, regardless of OS.)
What it all comes down to is that the user/super-user separation really does not provide any significant protections against viruses, especially on typical desktop systems. Sorry, but Linux is, for most intents and purposes, just as vulnerable to these types of viruses as Windows is.
Unix's security model is far from ideal. It's a very simple model that can't really do a whole lot. Not that Windows is any better; in my book, Windows is just a colorful variant of Unix. On the other hand, an OS that supports capability-based security (like EROS) would actually be able to safely run untrusted software -- viruses and all -- without harming the system, or even the individual user running it. Sadly, the idea has not been implemented in any mainstream OS (though I am currently working on a project that would bring such ideas to existing OS's).
I really wish people would stop making OS's that just copy Unix and create something new already! There are so many great ideas out there; so many better paradigms. Ugh...
As a resident of the state of Minnesota, I can tell you that Norm Coleman is a slimeball. If he seems to be saying or doing anything that sounds good to you, he's probably just doing it to get people to like him. Not that I think Mondale would have done any better. I voted for the third-party guy who actually seemed to have a clue. Why do the people who have clues always have to be from the third parties who never win?
Anyway, if this effort actually succeeds in its goals, great, but I suspect it will just be "for show" and "lip service" as you say. Sorry.
I'm confused. Are you saying that Microsoft's POSIX layer has no multithreading? Because not only does the article say otherwise, but it says right there in the writeup:
It contains... POSIX libraries and utilities including pthreads.
That's right... By some freak coincidence, KISS had developed a custom subtitle format and gave it the name "MPSub". You might think that the "MP" in that name stood for MPlayer -- you know, as if MPlayer had invented it -- but you would be wrong.
Amazon.com has had Doom 3 for pre-order for quite some time. They list the release date as March 1, 2004. Where did they get that date, you ask? They made it up!
id has a very specific policy regarding release dates: There are none. They will release the game when it is done, but until that point, any date you see is pure speculation.
At one point, Amazon was actually offering a discount if you bought Doom 3 together with Half-Life 2, which I thought was pretty amusing.
Admittedly, I don't keep up-to-date on any of the software in these tests other than the games, but... Has *any* of this software been ported to 64-bit? You can't just take any old 32-bit program, run it on a 64-bit processor, and say "hey, it's 64-bit!". The software has to be compiled for the 64-bit instruction set, and often requires source code changes as well. I'm pretty damned sure that none of those games have been ported to 64-bit, and I highly doubt any of the other software has been.
So, basically, all this test is doing is comparing the 32-bit speeds of these processors. 32-bit programs have nothing to gain from a 64-bit architecture (as far as I know). So, there's no reason to hype this as the battle of the 64-bit processors, because we haven't even seen their 64-bit speeds yet.
I suppose Microsoft likes to fall back in their improvements then, given that Windows XP DOES slow down over time, and that Windows XP is their latest OS.
Honestly, I haven't used XP, so I'm not going to pretend I know. However, the "Windows Rot" effect was a hallmark of the 9x series. I have never heard of the NT kernel having any such problems, and Win2k certainly hasn't had such problems for me, but whatever. I suppose it's remotely possible that XP took a huge step backwards. (Based on your next statement, though, I am inclined to believe that you have no idea what you're talking about.)
Even if they didn't, a simple program that crashes can bring down your entire system, as windows doesn't have protected memory.
This is absolutely, positively, 100% FALSE. Not only is it false, but it is an absurd claim. Windows most certainly does use protected memory. Even Windows 9x used protected memory. The last release of Windows to not use protected memory was 3.1 (almost a decade ago). The very first version of Windows NT used protected memory.
Hell, protected memory is so commonplace that people don't bother mentioning it anymore. It is simply assumed that any modern OS uses protected memory, because otherwise it would be an embarassment to the people who created the OS. It is only in the Mac world than you hear the term "protected memory" used in advertising because it is only in the Mac world where this is a new thing.
I suppose next you'll tell me that Windows doesn't have preemtive multitasking.
I said exponentially because the harder the work, the bigger this difference in time will be.
"Exponentially" suggests that the harder the work, the greater the percentage difference in time will be. You're suggesting that, if writing and sending a single e-mail is twice as fast on OSX, then writing and sending 10 e-mails is 1024 times as fast, and writing and sending 20 e-mails is 1048576 times as fast. I don't think that's the case. Maybe I'm just being pedantic here, though. I'm a programmer, so "exponentially faster" has a very specific meaning to me.
Besides that, I completely disagree with your statement. I never feel that I have to work to make my system run well. Win2k never crashes on me, never rots, never needs to be rebooted, etc. It's the software I use which determines my efficiency, and the software I use works great for what I do. OSX does not make me any more efficient. (I own both a Mac running OSX and a PC running Win2k, so I say this based on experience.)
Wow... Either you are a troll and you're just saying these things to invoke flame, or you are completely clueless.
Please keep in mind that now that Mac OS X is out, the speed of the Mac OS system does not degrade over time as Windows still does.
Having had my Win2k system up for over two months at a time, playing games frequently, I can pretty much assure you that Windows no longer degrades over time either. You are thinking of Win9x, which has been discontinued.
Simple tasks like opening docs, moving files, etc can still crash XP for instance.
No, no they can not. Not unless your hardware is broken.
The time it gets to get a job done in Mac OS X is exponentially lower than it is in Windows XP.
You clearly don't know what the term "exponentially" means.
Like I said, I personally believe that OSX is a much better OS than Windows XP, and that typical Mac hardware is far superior to typical PC hardware, but the kind of stuff you are saying is pure BS.
Your logic is as follows: "Apple does things to improve their performance that everyone else does too, therefore Apple does more than everyone else." This is not a sound argument. Why should we believe that just because they do stuff everyone else does, that they also do things that other companies don't do?
Besides, it is IBM that is improving the PPC, not Apple. No, IBM is not doing it specifically for Apple, nor are they doing it because Apple asked them to. IBM is also improving the PPC for the sake of, for example, Nintendo (Game Cubes use PPC processors). Sorry, but all Apple is doing here is sitting by and saying "Yay! Faster processors!".
Now, I am a big Apple fan myself; not so much for speed as for the fact that their products are so elegantly designed at every level, from the processor to the hardware layout to the operating system to the applications. However, your argument is still bogus.
I wish hardware manufacturers would learn that they have nothing to lose by releasing the specs on the system.
What on Earth makes you think that?
ATI would LOVE to know how NVidia's graphics cards work, and vice versa. Graphics cards are very complicated pieces of hardware that require far more research and development than just about anything else in your computer (perhaps even more than the CPU itself). The interface between the card and the OS is not by any means simple, and designing a good interface can make quite a difference in the speed and stability of the card. If NVidia released their specs, not only would ATI be able to get some insight out of it, but other companies who aren't quite so advanced (like S3) could simply read the spec and skip a very significant portion of the design process for their next card.
Furthermore, releasing the spec isn't going to do as much good as you think. Someone would still have to write the drivers. Now, open source has worked great for other kinds of device drivers, but graphics cards are entirely different. Graphics drivers have hundreds of times the functionality of almost any other sort of device driver. OpenGL is a huge library with hundreds (if not thousands) of functions, and all of the extensions supported by modern graphics cards can easily double that size. Furthermore, the OpenGL specification requires that any GL functionality not implemented in hardware be done in software. NVidia has spent a lot of time optimizing their drivers, including software implementations of a large chunk of OpenGL. They have a team of tens if not hundreds of programmers working on it full-time, and those programmers have the luxury of being able to walk down the hall and talk to the people who designed the hardware if they can't figure something out. I like open source at all (I run several open source projects), but I really don't think an open source driver would ever match what NVidia has produced.
And what about releasing the driver open source? Well, remember how I said that OpenGL functionality that isn't supported in hardware must be done in software? NVidia's software implementations of GL and their whole driver infrastructure could easily be used in competitors' products, saving that competitor millions of dollars of R&D.
Bottom line: If NVidia were to release hardware specs or open source drivers, they would lose a lot of money. It has nothing to do with piracy.
(Side Note: The Linux kernel maintainers have been known to frequently change the kernel interfaces in ways that break closed-source drivers, even though it is often possible for them to avoid such changes. As a result, NVidia is forced to spend a lot of effort updating their drivers for the latest kernel modifications, which they do even though Linux already represents a miniscule portion of their market. Maybe you should let the kernel devs know that you'd like to see better binary compatibility between kernel versions, so that your graphics drivers aren't always broken.)
(Oh, and, as already noted, ATI isn't doing any better as far as Linux goes.)
[~] echo 'void main(){printf("Hello World!\n");}' > hello.c [~] gcc hello.c -o hello hello.c: In function `main': hello.c:1: warning: return type of `main' is not `int' [~]./hello Hello World! [~]
LOL! Damnit, I thought I had fixed those, but then I had to undo a bunch of changes to get around the lameness filter, and I guess I accidentally undid that too.
Slashdot Mirror
I wonder why this was modded "troll"? "Redundant" I could see, but otherwise everything in it is quite true and well-reasoned. Was it because of your user name and sig? Or was it because you expressed a preference for Windows over Linux? (I found that hillarious, BTW, and completely agree with the sentiment. Of course, to each his own and all that.)
I am reading the patent, and I really don't understand how this could even be interpreted as being a patent on the concept of XML. It is a patent on a system that uses XML, as is pretty explicitly stated in the abstract.
Whether the patent itself is overly broad is up for debate. However, you can't just quote one line from the abstract and claim that the patent applies to everything in the universe that fits that one line. There is a reason for the body of the patent: to describe the specifics of the invention they are patenting.
Why would they build it off the Linux kernel when they already own BeOS? Yeah, I'd say that's a rumor alright.
Try searching for "sco". You get both.
Google blocked the "litigious bastards"->sco.com googlebomb. Why? Probably because, even though they hate SCO, they'd like to do whatever they can to avoid a legal fight.
Most googlebombs work within 48 hours. Several days after this one was suggested, it still wasn't working. At that point, I (yes, I myself) went around to a bunch of sites and suggested they change the link to caldera.com. Most of them ignored me, but I guess enough switched to make it work. Yay!
No no, you misread. He didn't say 50%. He said .50%. That comes to $312.50 a year. That's probably a lot more than what the guy makes in his parents' basement, though.
As a server, Linux is more stable. As a desktop, Windows (2k or XP, not 9x) is more stable. This is why I use Linux on my servers and Windows XP on my desktop. (However, I intend to switch to FreeBSD on the servers and Mac OSX on the desktop when I get the chance.)
X is slow sometimes, and crashes to often at my site (twice a week)
Impossible! Nothing unix-based ever crashes! You must be thinking of Windows.
(Running a Stable Hardware Platform)
Bull. The fact that XP froze on you proves only that you are not running a stable platform. Try ASUS A7N8X mobo + NVidia reference video with adequate cooling and you'll never lock up, period. At least, that's what I have, and XP never locks on me. And no, it's not that I don't stress it. I run games galore, develop software, run servers, use two monitors, have two RAID arrays installed (a RAID-5 and a RAID-0), etc. etc.
(And if Linux is perfectly stable for you, all that proves is that Linux happens to have better drivers for your particular hardware, or perhaps drivers that use fewer hardware features. In any case, that's the hardware manufacturer's fault, not XP's.)
Bad apps do not crash XP. Bad hardware and drivers crash XP. You must be thinking of 9x.
First, you'd have to save it to your hard drive, clicking on it wouldn't work
Most Windows e-mail clients will not open an executable when you simply click on it. In fact, they usually open multiple warning windows saying, essentially, "If you run this, you are a complete and total moron. Are you a moron? [YES] [NO]".
Then you'd need to "chmod +x" it
This provides about the same amount of protection as said warning windows. In order to run the program, you have to be fully aware that you are trying to run an executable. Having to chmod +x it is just an inconvenience, really.
and then you could run it as your user, in which case it can infect only things associated with that user.
Seriously, how many people read their e-mail on multi-user machines? Yes, I know there are some, but it's rare. In most cases, the person reading the e-mail is the only user of the particular machine they are on, and so having their own account totally trashed isn't really any better than just having the whole computer trashed.
Besides that, most viruses these days can accomplish all of their goals just as easily from a user account as they can from root. Typically, this involves propogating itself (requires only network access) and then carrying out some form of DDoS (again, requires only network access). Who needs root?
(Of course, on Windows, if you're smart, you're probably running ZoneAlarm, which will tell you when a program tries to access the internet and allow you to deny it that access. I am not familiar with any similar software for Linux. Though, if you're smart, you aren't running attachments anyway, regardless of OS.)
What it all comes down to is that the user/super-user separation really does not provide any significant protections against viruses, especially on typical desktop systems. Sorry, but Linux is, for most intents and purposes, just as vulnerable to these types of viruses as Windows is.
Unix's security model is far from ideal. It's a very simple model that can't really do a whole lot. Not that Windows is any better; in my book, Windows is just a colorful variant of Unix. On the other hand, an OS that supports capability-based security (like EROS) would actually be able to safely run untrusted software -- viruses and all -- without harming the system, or even the individual user running it. Sadly, the idea has not been implemented in any mainstream OS (though I am currently working on a project that would bring such ideas to existing OS's).
I really wish people would stop making OS's that just copy Unix and create something new already! There are so many great ideas out there; so many better paradigms. Ugh...
As a resident of the state of Minnesota, I can tell you that Norm Coleman is a slimeball. If he seems to be saying or doing anything that sounds good to you, he's probably just doing it to get people to like him. Not that I think Mondale would have done any better. I voted for the third-party guy who actually seemed to have a clue. Why do the people who have clues always have to be from the third parties who never win?
Anyway, if this effort actually succeeds in its goals, great, but I suspect it will just be "for show" and "lip service" as you say. Sorry.
According to the article, pthreads are a new feature in this version.
I'm confused. Are you saying that Microsoft's POSIX layer has no multithreading? Because not only does the article say otherwise, but it says right there in the writeup:
... POSIX libraries and utilities including pthreads.
It contains
That's right... By some freak coincidence, KISS had developed a custom subtitle format and gave it the name "MPSub". You might think that the "MP" in that name stood for MPlayer -- you know, as if MPlayer had invented it -- but you would be wrong.
uhh...
Amazon.com has had Doom 3 for pre-order for quite some time. They list the release date as March 1, 2004. Where did they get that date, you ask? They made it up!
id has a very specific policy regarding release dates: There are none. They will release the game when it is done, but until that point, any date you see is pure speculation.
At one point, Amazon was actually offering a discount if you bought Doom 3 together with Half-Life 2, which I thought was pretty amusing.
If not for the money we "wasted" going to the moon a few decades ago, you very well might not have had a computer with which to post that comment.
that song it uses as a beacon composed by the British band Blur
Maybe they didn't pick up Beagle's signal because it was indistinguishable from background noise?
Admittedly, I don't keep up-to-date on any of the software in these tests other than the games, but... Has *any* of this software been ported to 64-bit? You can't just take any old 32-bit program, run it on a 64-bit processor, and say "hey, it's 64-bit!". The software has to be compiled for the 64-bit instruction set, and often requires source code changes as well. I'm pretty damned sure that none of those games have been ported to 64-bit, and I highly doubt any of the other software has been.
So, basically, all this test is doing is comparing the 32-bit speeds of these processors. 32-bit programs have nothing to gain from a 64-bit architecture (as far as I know). So, there's no reason to hype this as the battle of the 64-bit processors, because we haven't even seen their 64-bit speeds yet.
I suppose Microsoft likes to fall back in their improvements then, given that Windows XP DOES slow down over time, and that Windows XP is their latest OS.
Honestly, I haven't used XP, so I'm not going to pretend I know. However, the "Windows Rot" effect was a hallmark of the 9x series. I have never heard of the NT kernel having any such problems, and Win2k certainly hasn't had such problems for me, but whatever. I suppose it's remotely possible that XP took a huge step backwards. (Based on your next statement, though, I am inclined to believe that you have no idea what you're talking about.)
Even if they didn't, a simple program that crashes can bring down your entire system, as windows doesn't have protected memory.
This is absolutely, positively, 100% FALSE. Not only is it false, but it is an absurd claim. Windows most certainly does use protected memory. Even Windows 9x used protected memory. The last release of Windows to not use protected memory was 3.1 (almost a decade ago). The very first version of Windows NT used protected memory.
Hell, protected memory is so commonplace that people don't bother mentioning it anymore. It is simply assumed that any modern OS uses protected memory, because otherwise it would be an embarassment to the people who created the OS. It is only in the Mac world than you hear the term "protected memory" used in advertising because it is only in the Mac world where this is a new thing.
I suppose next you'll tell me that Windows doesn't have preemtive multitasking.
I said exponentially because the harder the work, the bigger this difference in time will be.
"Exponentially" suggests that the harder the work, the greater the percentage difference in time will be. You're suggesting that, if writing and sending a single e-mail is twice as fast on OSX, then writing and sending 10 e-mails is 1024 times as fast, and writing and sending 20 e-mails is 1048576 times as fast. I don't think that's the case. Maybe I'm just being pedantic here, though. I'm a programmer, so "exponentially faster" has a very specific meaning to me.
Besides that, I completely disagree with your statement. I never feel that I have to work to make my system run well. Win2k never crashes on me, never rots, never needs to be rebooted, etc. It's the software I use which determines my efficiency, and the software I use works great for what I do. OSX does not make me any more efficient. (I own both a Mac running OSX and a PC running Win2k, so I say this based on experience.)
Wow... Either you are a troll and you're just saying these things to invoke flame, or you are completely clueless.
Please keep in mind that now that Mac OS X is out, the speed of the Mac OS system does not degrade over time as Windows still does.
Having had my Win2k system up for over two months at a time, playing games frequently, I can pretty much assure you that Windows no longer degrades over time either. You are thinking of Win9x, which has been discontinued.
Simple tasks like opening docs, moving files, etc can still crash XP for instance.
No, no they can not. Not unless your hardware is broken.
The time it gets to get a job done in Mac OS X is exponentially lower than it is in Windows XP.
You clearly don't know what the term "exponentially" means.
Like I said, I personally believe that OSX is a much better OS than Windows XP, and that typical Mac hardware is far superior to typical PC hardware, but the kind of stuff you are saying is pure BS.
Your logic is as follows: "Apple does things to improve their performance that everyone else does too, therefore Apple does more than everyone else." This is not a sound argument. Why should we believe that just because they do stuff everyone else does, that they also do things that other companies don't do?
Besides, it is IBM that is improving the PPC, not Apple. No, IBM is not doing it specifically for Apple, nor are they doing it because Apple asked them to. IBM is also improving the PPC for the sake of, for example, Nintendo (Game Cubes use PPC processors). Sorry, but all Apple is doing here is sitting by and saying "Yay! Faster processors!".
Now, I am a big Apple fan myself; not so much for speed as for the fact that their products are so elegantly designed at every level, from the processor to the hardware layout to the operating system to the applications. However, your argument is still bogus.
I wish hardware manufacturers would learn that they have nothing to lose by releasing the specs on the system.
What on Earth makes you think that?
ATI would LOVE to know how NVidia's graphics cards work, and vice versa. Graphics cards are very complicated pieces of hardware that require far more research and development than just about anything else in your computer (perhaps even more than the CPU itself). The interface between the card and the OS is not by any means simple, and designing a good interface can make quite a difference in the speed and stability of the card. If NVidia released their specs, not only would ATI be able to get some insight out of it, but other companies who aren't quite so advanced (like S3) could simply read the spec and skip a very significant portion of the design process for their next card.
Furthermore, releasing the spec isn't going to do as much good as you think. Someone would still have to write the drivers. Now, open source has worked great for other kinds of device drivers, but graphics cards are entirely different. Graphics drivers have hundreds of times the functionality of almost any other sort of device driver. OpenGL is a huge library with hundreds (if not thousands) of functions, and all of the extensions supported by modern graphics cards can easily double that size. Furthermore, the OpenGL specification requires that any GL functionality not implemented in hardware be done in software. NVidia has spent a lot of time optimizing their drivers, including software implementations of a large chunk of OpenGL. They have a team of tens if not hundreds of programmers working on it full-time, and those programmers have the luxury of being able to walk down the hall and talk to the people who designed the hardware if they can't figure something out. I like open source at all (I run several open source projects), but I really don't think an open source driver would ever match what NVidia has produced.
And what about releasing the driver open source? Well, remember how I said that OpenGL functionality that isn't supported in hardware must be done in software? NVidia's software implementations of GL and their whole driver infrastructure could easily be used in competitors' products, saving that competitor millions of dollars of R&D.
Bottom line: If NVidia were to release hardware specs or open source drivers, they would lose a lot of money. It has nothing to do with piracy.
(Side Note: The Linux kernel maintainers have been known to frequently change the kernel interfaces in ways that break closed-source drivers, even though it is often possible for them to avoid such changes. As a result, NVidia is forced to spend a lot of effort updating their drivers for the latest kernel modifications, which they do even though Linux already represents a miniscule portion of their market. Maybe you should let the kernel devs know that you'd like to see better binary compatibility between kernel versions, so that your graphics drivers aren't always broken.)
(Oh, and, as already noted, ATI isn't doing any better as far as Linux goes.)
Not to mention a preview button...
LOL! Damnit, I thought I had fixed those, but then I had to undo a bunch of changes to get around the lameness filter, and I guess I accidentally undid that too.