Is that linux developers simply 'roll back' to wherever Linux was before IBM first contributed, and start again from there. I doubt it would take much time to get back to where they where, especialy if programmers simply 'resubmitted' patches that they were sure they owned the rights to and had not been tainted by IBM.
Actually, not in this case (IANAL). They have indicated that their customers won't be held accountable to SCO... but that makes the code in question (all of it) a violation of the license under which it was distributed. Which means the users have NO right to run the code.
One of the first things it says is that you don't need to agree to the license to run the code, only to distribute the code.
SCO has no right to distribute Linux whatsoever, since they know 'their' code is in Linux without giving up the rights to that code. However, as long as they are not distributing any GPL'd code, they are in the clear.
Contrary to what a lot of people think, the worst that can happen to you if you violate the GPL is that you lose your rights to distribute the code, and you might have to pay a civil fine for the GPL infringement.
But the point is moot anyway, since SCO isn't distributing anything right now, they are not GPLing anything.
(Have you ever actually tried to port a J2EE app from one app server to another? Let alone one system architecture to another!?!)
I've never ever and a problem porting running java code on various platforms. I can imagine it could be a pain to move from one app server to another, but you can probably run that app server on any underlying CPU or OS. (depending on how much native code the app server uses, which shouldn't be very much)
Linux's many UI's could easily be described as overwhelming, and are definely inconsistent with each other. But I consider at least 3 UI's superior to OS X's Aqua. They are (in order or superiority); Enlightenment, gnome, and kde.
Of course you do. Some people just love screwing around with computers and this is just one more avenue. But some people just want their computers to work. And for these people Linux is never going to their first choice. OSX, however may very well be.
OSX and Linux target different niches. The reason that Linux killed those other unixes is because Linux could do the same thing that they did for free. (and lets not forget that it killed them in the office, where bottom line calculations are a lot more merciless)
Ok, I guess tell that to the CPU which does use ordered bits to address memory... bits do play into addressing and this is the main aspect of toting (by the vendors) 64 bit support in the newer processors.
Er, yeah I guess this was a misstatement of mine.. The 'bittness' of the CPU does play an important part in the addressing scheme, but not the only one. Generally how many 'bits' a CPU has is measured by the size of the integer used, the size of the general purpose registers, etc. Since pointers are generally generally ints, the bitness of a CPU has an affect on the way memory is addressed.
You cannot address more memory then you have bits for, so 32 bit systems (ones with VM systems that support 32 addressing) are locked at a per process limit of 4GBs (usually less because of shared/mapped memory regions). Sure some current so called 32b CPUs can address 36b or more of physical RAM (including the current G4), however this doesn't benefit processes unless the hardware wrapped around the CPU supports that additional bits and the OSes VM supports using them.
OSs like windows and Linux? And it is possible for applications which are specially written to take advantage of the larger memory sizes. I know Microsoft SQL server supports this, for example. Anyway, not many applications you need to run on a desktop PC need more per process memory then 4gb.
And yes, obviously more bits is better and can speed things up. Anyway, by now I've forgotten exactly what it was we started arguing about. Oh well.
Look, there were cheap alpha desktops for sale, and people used 'em. they were 64 bit desktops. There were computers that sat on desks that were 64 bit. That were personal, a computer that you sat at and used by yourself. It's idiotic to claim "the first 64 bit Personal Computer" when a) it's not true, and b) 64 bit technology has appeared in $90 game systems and PDAs years ago. I mean, fuck a lot of people take the term PC to mean IBM compatible anyway.
And it annoys me to have my intelegence insulted in that way.
The world's first 64-bit personal computer, with standard 3.5" hard drives, an optical drive, a 3-d video card, support for several GB of memory, and a whole bunch of other stuff in a tower enclosure, marketed at not only professionals but consumers who have some money to spend too.
Yeah, that just rolls off the tongue. Somebody call Apple and tell them to use this new product slogan!
How 'bout they just DON'T SAY IT? Is that so fucking hard?
There are far to many apple zealots out there. I mean seriously, these people ran the original Mac OS! About one of the crappiest OSs in the late 90s. I mean, no preemptive multitasking, no unthreading, barely any memory protection (guard pages only, and only in the very latest versions). I mean, good god
Plus, Linux has a long way to go before there is a consistent UI that comes anywhere near the quality of OSX (sorry to say:P). Sure, it looks nice these days, even pretty cool looking. But Aqua is just smooth like that.
Even if the hoard of Command line loving geeks is larger then the installed base of Mac users, I don't think Apple has much to worry about. The windows installed base was much larger then the apple and even when the OS was far crappier (since win-95 until OSX) and even when the hardware was slower and more expensive they stuck by their platform.
They are brainwashed. Nothing is going to change that:P
Could you ever walk down to a retail store and buy an Alpha for $2K? And run MS Office or a shrinkwrap game on it? Apple's claim seems pretty reasonable to me.
No, but I only paid $350 for my MIPS r4300 based PDA. It came with pocket office, and there were a bunch of wince games to play on it, and there were emulators for NES and SNES to play any of the games for those systems.
With that in mind, I don't see any trickery whatsoever. No previous 64-bit machine was ever considered a PERSONAL computer. Alpha, SGI, Sun, HP-PA, Itanium, even Opteron. Calm down everyone!
Considered by whom? My the mips r4300 based PDA I got 3 years ago seemed pretty personal to me.
A desktop computer is a computer that you can put on your desk. There were lots of alpha workstations that fit on desks.
How can you not see the difference between an Alpha and a PowerMac?
It's not any greater then the difference between the PowerMac and a standard PC.
Have there been 64-bit Windows PCs running around for years? No.
No, but there have been 64-bit linux PCs running on alpha for years. And they were pretty popular from what I remember. There was also a port of windows NT to alpha, but I'm not sure it ran in 64bit mode.
This isn't subversion of technical facts -- it's simple marketing. It's not blatant lies, it's little white ones that don't matter to most consumers. There are only so many ways you can say "This machine is very fast."
If they didn't matter, then there would be no reason for apple to say them. On the other hand, it means that there are going to be apple zealots talking about how great apple is because they brought the first 64 bit desktop to the world, which will be annoying.
The ps2 had been released before apple started up with those ads.
No, exactly the same. If the processor can't address more than 4 GB of physical RAM, then the kernel can't slice that RAM up into chunks and hand it out to the various processes. If you have 8 GB of RAM in a 32-bit machine, you're wasting 4 GB of RAM.
That's ridiculous. The 32bit Intel chips can access way more then 4gigs of ram. They use the same segment+offset system that allowed the 286 to access more then 64k of ram. Modern Xeon chips can access 2^36 bytes of ram, or 64 gigabytes. Do you really think 8-bit CPUs could only use 256 bytes of memory?
Second of all, even if a CPU can't access that much memory directly, they can use a paging system to flip between different 'pages' of memory. This was popular back in the 8/16 bit CPU days. These days, CPUs use the virtual memory system to give each application it's own address space, which maps to a real address space in hardware. So even if a program isn't written to take advantage of > 4 gig address, other programs on the system can use that space.
Seriously now... 64b addressing isn't about speed its about being able to address lots of memory. Many professional, engineering, scientific and even some prosumer task can easily use more then 4GB of memory, even greater then 8GB of memory (current system limit in the new PowerMac G5s).
How many 'bits' a CPU has isn't about addressing at all. AMD's new chips can't address a full 64 bits of space yet, and older Intel chips could address 36 bits. I'm pretty sure the current crop of xeon CPUs can do this.
No the performance difference comes when you have to deal with integers larger then 2^[num bits]. If you have a 16 bit CPU, in order to add two 32 bit numbers together, you need to add the first half, then add the second half, then add the carries from the first half. It takes way more clock cycles then it would for a 32bit CPU. And you often need to deal with numbers larger then 64k. On the other hand, you don't often need to deal with integers larger then 4.3 billion. All of the major computers these days support huge-ass floating points, so the need for 64 bit precision isn't as much. It certainly can help, but it isn't as much as the performance boost from 8 to 16, or 16 to 32.
Indeedy. Kind of amusing to note how it "breaks through the 4gig barrier".
I've got 1.5 gigs of ram in my machine, and a self-coded app that takes shitload (I could have spent hours upon hours getting the mem use down, or I could have purchased more ram for insanely low prices). Memory is so cheap now that putting 4gigs of ram in your PC isn't even impractical.
The reason it is a stupid comment though is that high end Intel chips have supported more then 4 gigs for quite a while, using the same funky segments+offsets system they used back in the 286 days to allow them to access more then 64k. And not only that, the new g5 machines only support 8 gigs of ram anyway, so it's not really that much of an improvement.
Indeedy. Kind of amusing to note how it "breaks through the 4gig barrier".
The reason it is a stupid comment though is that high end Intel chips have supported more then 4 gigs for quite a while, using the same funky segments+offsets system they used back in the 286 days to allow them to access more then 64k. And not only that, the new g5 machines only support 8 gigs of ram anyway, so it's not really that much of an improvement.
Well, I've had some experience with business software, working for Iowa Student Loan Liquidity corp. On various projects. I can tell you first hand that almost everything we did sucked. The solutions we had for various problems were incredibly baroque and stupid.
The basic problem was, most of the people there were idiots. Can't really beat around the bush, the people were morons. Now I'm sure that at some corporations it's different, if the CIO or high-ups really know their stuff, but from what I've heard a lot of times they don't. Non technical people have no way to judge the ability of programmers, so basically you're stuck with the luck of the draw. Either you end up with smart people on top, or you have idiots all the way down. In the later case, any smart hires arenâ(TM)t appreciated for their brains.
Of course, you might read this as just another self-centered geek rant about the idiocy of the rest of the world, and there's a good chance you might be right, but those are my impressions anyway. (and the place I work now seems to have lots of smart people, and we don't do bussness logic)
It would be a win-win situation for them, and they can use their trademark to protect the 'purity'. I.E. if it's not "Pure java" it can't be called "Pure java". And microsoft seems to have gotten out of the java game anyway, so their corruption isn't much of an issue. I doubt the open source maintainers would allow contributions that would violate sun's standards, and Microsoft would never fork a GPL project since they hate the GPL so much.
And plus, sun wouldn't need to do any of the work themselves:P
The new G5 machines, with the IBM 970 processor, use the "world's first 64-bit desktop processor" (and the "fastest 64-bit processor ever")
Wow, I'm sure people who had Alpha workstations back in the day will be surprised. Even the n64 had a 64 bit processor, the MIPS r4300. The chip was $35 dollars in bulk in 1996. Iâ(TM)m pretty sure this chip has been used in PDAs in the past few years.
The only reason that they havenâ(TM)t been used in desktops so far is that A) There is a huge legacy base to support and B) The speed increase isn't even that great. I mean, you don't need more then 32 bit ints for the vast majority of the calculations you need to do on a PC (whereas on a 16 bit computer, you need to use several instructions to calculate 'both halves' of the number anytime you needed to do math with numbers larger then 64k.). And anyway, all of the major CPUs available today have instructions that deal with huge amounts of data for floating point and SIMD multimedia stuff.
I'm suppressed apple isn't claming that their machines do 'twice as much work' because they have twice as many bits. This subversion of technical facts for marketing purposes is something apple is constantly guilty of, and it's really annoying. Because you know you're going to have some idiot mac zealot come back at you with something like "yeah, well this is the first 64-bit desktop EVAR" Just like how they claimed the g4 was the first "Desktop supercomputer" or something like that, because it met some obsolete government export restrictions, the same restrictions that the playstation two had surpassed months before.
Slashdot Mirror
Is that linux developers simply 'roll back' to wherever Linux was before IBM first contributed, and start again from there. I doubt it would take much time to get back to where they where, especialy if programmers simply 'resubmitted' patches that they were sure they owned the rights to and had not been tainted by IBM.
Actually, not in this case (IANAL). They have indicated that their customers won't be held accountable to SCO... but that makes the code in question (all of it) a violation of the license under which it was distributed. Which means the users have NO right to run the code.
One of the first things it says is that you don't need to agree to the license to run the code, only to distribute the code.
Even microsoft can run GPL'd code.
SCO has no right to distribute Linux whatsoever, since they know 'their' code is in Linux without giving up the rights to that code. However, as long as they are not distributing any GPL'd code, they are in the clear.
Contrary to what a lot of people think, the worst that can happen to you if you violate the GPL is that you lose your rights to distribute the code, and you might have to pay a civil fine for the GPL infringement.
But the point is moot anyway, since SCO isn't distributing anything right now, they are not GPLing anything.
When did the taiwanese start caring about Intelectual property?
(Have you ever actually tried to port a J2EE app from one app server to another? Let alone one system architecture to another!?!)
I've never ever and a problem porting running java code on various platforms. I can imagine it could be a pain to move from one app server to another, but you can probably run that app server on any underlying CPU or OS. (depending on how much native code the app server uses, which shouldn't be very much)
Linux's many UI's could easily be described as overwhelming, and are definely inconsistent with each other. But I consider at least 3 UI's superior to OS X's Aqua. They are (in order or superiority); Enlightenment, gnome, and kde.
Of course you do. Some people just love screwing around with computers and this is just one more avenue. But some people just want their computers to work. And for these people Linux is never going to their first choice. OSX, however may very well be.
OSX and Linux target different niches. The reason that Linux killed those other unixes is because Linux could do the same thing that they did for free. (and lets not forget that it killed them in the office, where bottom line calculations are a lot more merciless)
Ok, I guess tell that to the CPU which does use ordered bits to address memory... bits do play into addressing and this is the main aspect of toting (by the vendors) 64 bit support in the newer processors.
Er, yeah I guess this was a misstatement of mine.. The 'bittness' of the CPU does play an important part in the addressing scheme, but not the only one. Generally how many 'bits' a CPU has is measured by the size of the integer used, the size of the general purpose registers, etc. Since pointers are generally generally ints, the bitness of a CPU has an affect on the way memory is addressed.
You cannot address more memory then you have bits for, so 32 bit systems (ones with VM systems that support 32 addressing) are locked at a per process limit of 4GBs (usually less because of shared/mapped memory regions). Sure some current so called 32b CPUs can address 36b or more of physical RAM (including the current G4), however this doesn't benefit processes unless the hardware wrapped around the CPU supports that additional bits and the OSes VM supports using them.
OSs like windows and Linux? And it is possible for applications which are specially written to take advantage of the larger memory sizes. I know Microsoft SQL server supports this, for example. Anyway, not many applications you need to run on a desktop PC need more per process memory then 4gb.
And yes, obviously more bits is better and can speed things up. Anyway, by now I've forgotten exactly what it was we started arguing about. Oh well.
Look, there were cheap alpha desktops for sale, and people used 'em. they were 64 bit desktops. There were computers that sat on desks that were 64 bit. That were personal, a computer that you sat at and used by yourself. It's idiotic to claim "the first 64 bit Personal Computer" when a) it's not true, and b) 64 bit technology has appeared in $90 game systems and PDAs years ago. I mean, fuck a lot of people take the term PC to mean IBM compatible anyway.
And it annoys me to have my intelegence insulted in that way.
The world's first 64-bit personal computer, with standard 3.5" hard drives, an optical drive, a 3-d video card, support for several GB of memory, and a whole bunch of other stuff in a tower enclosure, marketed at not only professionals but consumers who have some money to spend too.
Yeah, that just rolls off the tongue. Somebody call Apple and tell them to use this new product slogan!
How 'bout they just DON'T SAY IT? Is that so fucking hard?
There are far to many apple zealots out there. I mean seriously, these people ran the original Mac OS! About one of the crappiest OSs in the late 90s. I mean, no preemptive multitasking, no unthreading, barely any memory protection (guard pages only, and only in the very latest versions). I mean, good god
:P). Sure, it looks nice these days, even pretty cool looking. But Aqua is just smooth like that.
:P
Plus, Linux has a long way to go before there is a consistent UI that comes anywhere near the quality of OSX (sorry to say
Even if the hoard of Command line loving geeks is larger then the installed base of Mac users, I don't think Apple has much to worry about. The windows installed base was much larger then the apple and even when the OS was far crappier (since win-95 until OSX) and even when the hardware was slower and more expensive they stuck by their platform.
They are brainwashed. Nothing is going to change that
Of course, it would have taken less time for you, since you guys already knew the codebase, but I digress.
Explain whats wrong or fuck off, idiot.
I meant "yes, also..." rather then "no, but..."
Could you ever walk down to a retail store and buy an Alpha for $2K? And run MS Office or a shrinkwrap game on it? Apple's claim seems pretty reasonable to me.
No, but I only paid $350 for my MIPS r4300 based PDA. It came with pocket office, and there were a bunch of wince games to play on it, and there were emulators for NES and SNES to play any of the games for those systems.
With that in mind, I don't see any trickery whatsoever. No previous 64-bit machine was ever considered a PERSONAL computer. Alpha, SGI, Sun, HP-PA, Itanium, even Opteron. Calm down everyone!
Considered by whom? My the mips r4300 based PDA I got 3 years ago seemed pretty personal to me.
Alpha workstations were just that, workstations.
A desktop computer is a computer that you can put on your desk. There were lots of alpha workstations that fit on desks.
How can you not see the difference between an Alpha and a PowerMac?
It's not any greater then the difference between the PowerMac and a standard PC.
Have there been 64-bit Windows PCs running around for years? No.
No, but there have been 64-bit linux PCs running on alpha for years. And they were pretty popular from what I remember. There was also a port of windows NT to alpha, but I'm not sure it ran in 64bit mode.
This isn't subversion of technical facts -- it's simple marketing. It's not blatant lies, it's little white ones that don't matter to most consumers. There are only so many ways you can say "This machine is very fast."
If they didn't matter, then there would be no reason for apple to say them. On the other hand, it means that there are going to be apple zealots talking about how great apple is because they brought the first 64 bit desktop to the world, which will be annoying.
The ps2 had been released before apple started up with those ads.
Are you actually criticizing Apple for not subverting technical facts for marketing purposes? Does that seem a little foolish to anyone else?
No, I'm criticizing them because they are doing that.
No, exactly the same. If the processor can't address more than 4 GB of physical RAM, then the kernel can't slice that RAM up into chunks and hand it out to the various processes. If you have 8 GB of RAM in a 32-bit machine, you're wasting 4 GB of RAM.
That's ridiculous. The 32bit Intel chips can access way more then 4gigs of ram. They use the same segment+offset system that allowed the 286 to access more then 64k of ram. Modern Xeon chips can access 2^36 bytes of ram, or 64 gigabytes. Do you really think 8-bit CPUs could only use 256 bytes of memory?
Second of all, even if a CPU can't access that much memory directly, they can use a paging system to flip between different 'pages' of memory. This was popular back in the 8/16 bit CPU days. These days, CPUs use the virtual memory system to give each application it's own address space, which maps to a real address space in hardware. So even if a program isn't written to take advantage of > 4 gig address, other programs on the system can use that space.
Seriously now... 64b addressing isn't about speed its about being able to address lots of memory. Many professional, engineering, scientific and even some prosumer task can easily use more then 4GB of memory, even greater then 8GB of memory (current system limit in the new PowerMac G5s).
How many 'bits' a CPU has isn't about addressing at all. AMD's new chips can't address a full 64 bits of space yet, and older Intel chips could address 36 bits. I'm pretty sure the current crop of xeon CPUs can do this.
No the performance difference comes when you have to deal with integers larger then 2^[num bits]. If you have a 16 bit CPU, in order to add two 32 bit numbers together, you need to add the first half, then add the second half, then add the carries from the first half. It takes way more clock cycles then it would for a 32bit CPU. And you often need to deal with numbers larger then 64k. On the other hand, you don't often need to deal with integers larger then 4.3 billion. All of the major computers these days support huge-ass floating points, so the need for 64 bit precision isn't as much. It certainly can help, but it isn't as much as the performance boost from 8 to 16, or 16 to 32.
Sorry, malformated HTML :P
Indeedy. Kind of amusing to note how it "breaks through the 4gig barrier".
I've got 1.5 gigs of ram in my machine, and a self-coded app that takes shitload (I could have spent hours upon hours getting the mem use down, or I could have purchased more ram for insanely low prices). Memory is so cheap now that putting 4gigs of ram in your PC isn't even impractical.
The reason it is a stupid comment though is that high end Intel chips have supported more then 4 gigs for quite a while, using the same funky segments+offsets system they used back in the 286 days to allow them to access more then 64k. And not only that, the new g5 machines only support 8 gigs of ram anyway, so it's not really that much of an improvement.
Indeedy. Kind of amusing to note how it "breaks through the 4gig barrier".
The reason it is a stupid comment though is that high end Intel chips have supported more then 4 gigs for quite a while, using the same funky segments+offsets system they used back in the 286 days to allow them to access more then 64k. And not only that, the new g5 machines only support 8 gigs of ram anyway, so it's not really that much of an improvement.
Well, I've had some experience with business software, working for Iowa Student Loan Liquidity corp. On various projects. I can tell you first hand that almost everything we did sucked. The solutions we had for various problems were incredibly baroque and stupid.
The basic problem was, most of the people there were idiots. Can't really beat around the bush, the people were morons. Now I'm sure that at some corporations it's different, if the CIO or high-ups really know their stuff, but from what I've heard a lot of times they don't. Non technical people have no way to judge the ability of programmers, so basically you're stuck with the luck of the draw. Either you end up with smart people on top, or you have idiots all the way down. In the later case, any smart hires arenâ(TM)t appreciated for their brains.
Of course, you might read this as just another self-centered geek rant about the idiocy of the rest of the world, and there's a good chance you might be right, but those are my impressions anyway. (and the place I work now seems to have lots of smart people, and we don't do bussness logic)
It would be a win-win situation for them, and they can use their trademark to protect the 'purity'. I.E. if it's not "Pure java" it can't be called "Pure java". And microsoft seems to have gotten out of the java game anyway, so their corruption isn't much of an issue. I doubt the open source maintainers would allow contributions that would violate sun's standards, and Microsoft would never fork a GPL project since they hate the GPL so much.
:P
And plus, sun wouldn't need to do any of the work themselves
How much you want to be they'll be outperformed by the p5 and Athlon64 by then?
The new G5 machines, with the IBM 970 processor, use the "world's first 64-bit desktop processor" (and the "fastest 64-bit processor ever")
Wow, I'm sure people who had Alpha workstations back in the day will be surprised. Even the n64 had a 64 bit processor, the MIPS r4300. The chip was $35 dollars in bulk in 1996. Iâ(TM)m pretty sure this chip has been used in PDAs in the past few years.
The only reason that they havenâ(TM)t been used in desktops so far is that A) There is a huge legacy base to support and B) The speed increase isn't even that great. I mean, you don't need more then 32 bit ints for the vast majority of the calculations you need to do on a PC (whereas on a 16 bit computer, you need to use several instructions to calculate 'both halves' of the number anytime you needed to do math with numbers larger then 64k.). And anyway, all of the major CPUs available today have instructions that deal with huge amounts of data for floating point and SIMD multimedia stuff.
I'm suppressed apple isn't claming that their machines do 'twice as much work' because they have twice as many bits. This subversion of technical facts for marketing purposes is something apple is constantly guilty of, and it's really annoying. Because you know you're going to have some idiot mac zealot come back at you with something like "yeah, well this is the first 64-bit desktop EVAR" Just like how they claimed the g4 was the first "Desktop supercomputer" or something like that, because it met some obsolete government export restrictions, the same restrictions that the playstation two had surpassed months before.
What's illegal filesharing? How can it be illegal to share files? It's not even illegal to violate copyright.