Will people jump off the upgrade treadmill and simply wait until their current machine dies before purchasing a new one?
I believe you'll find that 95% of people do that already. I'm still trying to decide how much longer I can live with my 2003-era Windows PC (with 2005 graphics card) before I have to give in and buy another one for gaming, and my laptop is literally dying which is the only reason why I'll replace it in the next year or so.
The difference that cheaper PCs has made to me is that I buy more of them for specific uses, rather than trying to do everything on one expensive system. Netbooks are a good example of that as people who wouldn't have paid $1000 for a laptop a few years back will pay $300 for a netbook.
Also, they boot a whole lot slower with the 160gig drive than they do with a 8gig flash.
Bootchart claims about 35 seconds to boot my single-core Atom netbook with hard drive, which is about 15 seconds slower than the HTPC system with a dual-core Atom and SSD. If I boot it up to use for half an hour, that's about a 1% saving in time for a 95% loss of storage space... for me that's an easy trade.
I have one of the early EeePCs - I think it's the 900A - with a 4GB SSD and a 9 inch screen. It runs for at least 5 hours, and depending on the pants I wear it can fit into a cargo pocket. *That's* a netbook.
The EeePC I bought a few months ago has a 100+GB hard drive, 10 or 11 inch screen and runs for at least as long (the battery display claims 9 hours but I don't quite believe it). The only downside is that it barely fits into my jacket pocket, but I couldn't live with a screen any smaller than it has anyway.
Don't the new dual-core Atom systems use less power than the old single cores?
No, it doesn't. And I wish that people would quit claiming that it does. The minimum wage laws have little if any effect on the number of jobs or the standard of living.
Cool. So increase the minimum wage to $100 an hour and everyone will be rich.
The only time that the minimum wage has no impact on jobs is when people are already paid that much or more... in which case it's useless. Any time it pushes wages above market rates, is merely insures that people whose labour isn't worth that much will be unemployed all their life... in which case it's evil.
I remember seeing something about this on Discovery or History channel years ago and a quick search pointed me to British Airways Flight 9 on Wikipedia, all four engines FAILED!
BA Flight 9 flew through a concentrated ash cloud, and no-one is saying that aircraft should do so. But there's a level between that concentration and zero where the ash causes no significant impact on the engines, at which point it's safe to fly; more than that, there are higher levels where the engines will require increased maintenance but the airlines may be willing to pay that cost in order to keep the planes flying.
The idea that a tiny level of ash will cause an airliner to fall out of the sky is just silly, and while I'd agree that closing down European airspace for a brief period was justified, keeping it closed for days was certainly an overreaction by burrowcrats who were too scared to take the risk of letting planes fly.
RHEL has given me significantly more pain than Ubuntu. They have a bad habit of breaking things in updates.
Freaky. I've had far more pain from Ubuntu updates than RHEL/CentOS.... the last CentOS upgrade from 5.3 to 5.4 took about an hour with almost no intervention, whereas the Ubuntu upgrade from 9.04 to 9.10 took a day with lots of intervention before the system was working again.
That said, many of the things I had to fix up on Ubuntu were services I don't run on CentOS, such as MythTV and Zabbix.
We use RHEL/CentOS for a lot of servers, but while they're stable and reliable, they're also using old versions of a lot of packages which aren't compatible with the latest shiny things. So if you want to run SuperWhizzoWebService you may well have to either upgrade packages on your RHEL server to the latest versions (which is often a real pain) or just run a more 'bleeding edge' distribution.
I'm not a fan of Ubuntu on servers, but if it has to run shiny things and doesn't need to be up 24/7/365 then it may well be a good choice.
This story confirms one thing, only stupid people will use stupid site like Facebook, and share all details abouth themselves.
So long as you include the 'and share all details about themselves', I'd certainly agree: if there's something you don't want your boss or your relatives to find out then... duh... don't post it on the Internet. In my case that's easy because my boss and my relatives are on my Facebook friends lists so I know not to post anything they wouldn't want to see.
Of course, but it will be a question of how many layers can you go until things become unusable. At some point it wont be realistic.
So long as you have a few generations of hardware between the 'layered' emulators it shouldn't be a big deal; back in the 90s I remember running a DOS-based Sinclair Spectrum emulator on a PC Emulator on a Sparc laptop and the games ran at the same speed as the original Spectrum despite having two layers of emulation between them and the real CPU.
Of course emulation bugs will tend to accumulate through the layers, so if you have fifteen layers you may find that it doesn't run the way it's supposed to:).
...and if you want a really horrible assembly language, try MIPS.
Geez, kids today can't even handle something as mundane as delay slots... they were common on the RISC chips I programmed in the 90s to hide pipeline stalls.
My favorite horrible assembly language was the one (I forget which, Clipper maybe?) where there was no hardware support for pipeline stalls, so if you performed an operation in one instruction you had to ensure you didn't try to use the result until N instructions later or the CPU would silently give you garbage. Great fun to debug.
Could be just the boost they need for that uC to give it a legitimate competitive edge over ARM.
When I was working with ARMs a few years ago our whole system took less power when playing back 720P video than my Atom CPU alone takes when it's idle. Even my Ion system takes about 29W to play HD video whereas I believe our ARM-based system was more like 2W.
Atoms are great when you must have x86 compatibility at fairly low power, but simply aren't competitive in most of the areas where ARMs are used.
As far as I'm aware the Nano has a higher power consumption than an Atom, which is already too power-hungry for most markets where ARM is the processor of choice.
But if Apple do buy ARM, then ARM is dead; no other hardware manufacturer is going to want to allow Apple to control their products. So some other company will get rich designing a similar chip.
With my current AM3 socket, I can upgrade to a 6-core AMD chip with just a BIOS update. Why can't Intel do that?
They could. But since only a tiny fraction of people ever upgrade CPUs, there's no reason to cripple your CPUs with support for old chipsets when you can just release a new one; every current AMD CPU has to support DDR2 RAM as well as DDR3, for example, and there's some evidence that requirement is significantly affecting AMD's memory performance with DDR3.
It would be different if AMD had better CPUs than Intel, but since Intel's are the fastest right now you can either buy the fastest CPU with the appropriate motherboard from Intel, or you can cheap out sticking your fancy new 8-core AMD chip into a dual-channel DDR2 system while Intel can release their 8-core chip with a quad-channel DDR3 motherboard.
Heck, the last two motherboards I bought from Intel came with the CPU soldered in.
Or rather, it worked perfectly fine 66% of the time. Three tries with one failure.
Actually, it's 697.31965% fine, according to the FPU on an 'unlocked' core one of the other two CPUs.
The problem with his claim that the other two are 'perfectly fine' is that he has no idea whether the cores really work 'perfectly fine' without performing the same kind of low-level tests that the manufacturer would have performed before disabling them. Of course if he can live with, say, the FPU randomly producing incorrect results then that may not matter.
I used to work for a company which disabled components of chips which failed manufacturing tests and sold them as cheaper, less powerful units, and with our mix of products I don't believe we ever had to disable working hardware. I can't vouch for Intel or AMD.
I fail (and therein lies the problem perhaps;-) ) to understand how ActiveX is more dangerous than plug-ins.
While that's true to some extent, there are three common Firefox plugins, all of which have had major security holes: Java, Flash and Adobe PDF. Most people don't need Java or PDF plugins, but Flash is harder to get rid of.
There are about a bazillion ActiveX things and most of them probably have major security holes.
By the time you get around to setting up that Solaris server, Btrfs will have stabilized through 3-4 more mainline kernel releases.
Which means that about five years later it will be ready for production use:).
I am thinking of switching one of my non-vital Linux systems to btrfs before long to try it out, but the whole point of setting up a ZFS server would be for proven reliable storage.
Why would a 'weekend hacker' work on OpenSolaris when they could work on Linux instead?
That's the fundamental problem: OpenSolaris has user features that Linux doesn't -- assuming Oracle continue to support it I'm probably going to set up an OpenSolaris server in the next year or so because ZFS is better than anything Linux currently has -- but it doesn't really offer anything to the average 'weekend hacker' that Linux doesn't.
Even if it was made available under the GPL, I suspect most of the best code would be copied into Linux and then it would die off.
I'd say they're much closer in peril-level than the sting/gunshot example above. They both pose a threat to the liberties, freedoms, and lives of Americans.
Lives, perhaps, but exactly what threat does bin Laden pose to Americans' liberties and freedoms? Is he going to run for President in 2012?
All the post-9/11 impositions on American freedoms have come from the US government, not some crazy guy in a cave in Afghanistan. And most of them are things the government have wanted to do for years but had no excuse to impose before that point.
The last thing Linux needs is a set-in-stone kernel interface: 'backwards compatibility' is what has ensured that Windows remains a steaming pile of kludges and security holes as no old components can be thrown away.
I can only presume that you are actually Bill Gates and want to destroy Linux by forcing it to repeat Windows' mistakes.
Once a machine gets owned it's gone. Total wipe, reinstall from good backup. No matter what OS or even WIndows it is.
Joe Sixpack doesn't have a backup.
Also, Joe Sixpack probably don't have XP CDs, so he has to install from the 'recovery partition'; I wonder whether any rootkits are installing themselves into the recovery partition so they'll automatically be reinstalled if someone tries to wipe their system and reinstall from scratch?
Slashdot Mirror
Will people jump off the upgrade treadmill and simply wait until their current machine dies before purchasing a new one?
I believe you'll find that 95% of people do that already. I'm still trying to decide how much longer I can live with my 2003-era Windows PC (with 2005 graphics card) before I have to give in and buy another one for gaming, and my laptop is literally dying which is the only reason why I'll replace it in the next year or so.
The difference that cheaper PCs has made to me is that I buy more of them for specific uses, rather than trying to do everything on one expensive system. Netbooks are a good example of that as people who wouldn't have paid $1000 for a laptop a few years back will pay $300 for a netbook.
Also, they boot a whole lot slower with the 160gig drive than they do with a 8gig flash.
Bootchart claims about 35 seconds to boot my single-core Atom netbook with hard drive, which is about 15 seconds slower than the HTPC system with a dual-core Atom and SSD. If I boot it up to use for half an hour, that's about a 1% saving in time for a 95% loss of storage space... for me that's an easy trade.
I have one of the early EeePCs - I think it's the 900A - with a 4GB SSD and a 9 inch screen. It runs for at least 5 hours, and depending on the pants I wear it can fit into a cargo pocket. *That's* a netbook.
The EeePC I bought a few months ago has a 100+GB hard drive, 10 or 11 inch screen and runs for at least as long (the battery display claims 9 hours but I don't quite believe it). The only downside is that it barely fits into my jacket pocket, but I couldn't live with a screen any smaller than it has anyway.
Don't the new dual-core Atom systems use less power than the old single cores?
No, it doesn't. And I wish that people would quit claiming that it does. The minimum wage laws have little if any effect on the number of jobs or the standard of living.
Cool. So increase the minimum wage to $100 an hour and everyone will be rich.
The only time that the minimum wage has no impact on jobs is when people are already paid that much or more... in which case it's useless. Any time it pushes wages above market rates, is merely insures that people whose labour isn't worth that much will be unemployed all their life... in which case it's evil.
I remember seeing something about this on Discovery or History channel years ago and a quick search pointed me to British Airways Flight 9 on Wikipedia, all four engines FAILED!
BA Flight 9 flew through a concentrated ash cloud, and no-one is saying that aircraft should do so. But there's a level between that concentration and zero where the ash causes no significant impact on the engines, at which point it's safe to fly; more than that, there are higher levels where the engines will require increased maintenance but the airlines may be willing to pay that cost in order to keep the planes flying.
The idea that a tiny level of ash will cause an airliner to fall out of the sky is just silly, and while I'd agree that closing down European airspace for a brief period was justified, keeping it closed for days was certainly an overreaction by burrowcrats who were too scared to take the risk of letting planes fly.
RHEL has given me significantly more pain than Ubuntu. They have a bad habit of breaking things in updates.
Freaky. I've had far more pain from Ubuntu updates than RHEL/CentOS.... the last CentOS upgrade from 5.3 to 5.4 took about an hour with almost no intervention, whereas the Ubuntu upgrade from 9.04 to 9.10 took a day with lots of intervention before the system was working again.
That said, many of the things I had to fix up on Ubuntu were services I don't run on CentOS, such as MythTV and Zabbix.
We use RHEL/CentOS for a lot of servers, but while they're stable and reliable, they're also using old versions of a lot of packages which aren't compatible with the latest shiny things. So if you want to run SuperWhizzoWebService you may well have to either upgrade packages on your RHEL server to the latest versions (which is often a real pain) or just run a more 'bleeding edge' distribution.
I'm not a fan of Ubuntu on servers, but if it has to run shiny things and doesn't need to be up 24/7/365 then it may well be a good choice.
This story confirms one thing, only stupid people will use stupid site like Facebook, and share all details abouth themselves.
So long as you include the 'and share all details about themselves', I'd certainly agree: if there's something you don't want your boss or your relatives to find out then... duh... don't post it on the Internet. In my case that's easy because my boss and my relatives are on my Facebook friends lists so I know not to post anything they wouldn't want to see.
Of course, but it will be a question of how many layers can you go until things become unusable. At some point it wont be realistic.
So long as you have a few generations of hardware between the 'layered' emulators it shouldn't be a big deal; back in the 90s I remember running a DOS-based Sinclair Spectrum emulator on a PC Emulator on a Sparc laptop and the games ran at the same speed as the original Spectrum despite having two layers of emulation between them and the real CPU.
Of course emulation bugs will tend to accumulate through the layers, so if you have fifteen layers you may find that it doesn't run the way it's supposed to :).
software running without admin/root priviledges CANNOT break havoc in anything but the user account
If that user account is a POS terminal communicating credit card information to banks, controlling it could be just as bad as gaining root access.
...and if you want a really horrible assembly language, try MIPS.
Geez, kids today can't even handle something as mundane as delay slots... they were common on the RISC chips I programmed in the 90s to hide pipeline stalls.
My favorite horrible assembly language was the one (I forget which, Clipper maybe?) where there was no hardware support for pipeline stalls, so if you performed an operation in one instruction you had to ensure you didn't try to use the result until N instructions later or the CPU would silently give you garbage. Great fun to debug.
Could be just the boost they need for that uC to give it a legitimate competitive edge over ARM.
When I was working with ARMs a few years ago our whole system took less power when playing back 720P video than my Atom CPU alone takes when it's idle. Even my Ion system takes about 29W to play HD video whereas I believe our ARM-based system was more like 2W.
Atoms are great when you must have x86 compatibility at fairly low power, but simply aren't competitive in most of the areas where ARMs are used.
As far as I'm aware the Nano has a higher power consumption than an Atom, which is already too power-hungry for most markets where ARM is the processor of choice.
But if Apple do buy ARM, then ARM is dead; no other hardware manufacturer is going to want to allow Apple to control their products. So some other company will get rich designing a similar chip.
With my current AM3 socket, I can upgrade to a 6-core AMD chip with just a BIOS update. Why can't Intel do that?
They could. But since only a tiny fraction of people ever upgrade CPUs, there's no reason to cripple your CPUs with support for old chipsets when you can just release a new one; every current AMD CPU has to support DDR2 RAM as well as DDR3, for example, and there's some evidence that requirement is significantly affecting AMD's memory performance with DDR3.
It would be different if AMD had better CPUs than Intel, but since Intel's are the fastest right now you can either buy the fastest CPU with the appropriate motherboard from Intel, or you can cheap out sticking your fancy new 8-core AMD chip into a dual-channel DDR2 system while Intel can release their 8-core chip with a quad-channel DDR3 motherboard.
Heck, the last two motherboards I bought from Intel came with the CPU soldered in.
Or rather, it worked perfectly fine 66% of the time. Three tries with one failure.
Actually, it's 697.31965% fine, according to the FPU on an 'unlocked' core one of the other two CPUs.
The problem with his claim that the other two are 'perfectly fine' is that he has no idea whether the cores really work 'perfectly fine' without performing the same kind of low-level tests that the manufacturer would have performed before disabling them. Of course if he can live with, say, the FPU randomly producing incorrect results then that may not matter.
I used to work for a company which disabled components of chips which failed manufacturing tests and sold them as cheaper, less powerful units, and with our mix of products I don't believe we ever had to disable working hardware. I can't vouch for Intel or AMD.
Have fun: you don't need to click on anything to get owned by Flash malware served from an advertising site.
I fail (and therein lies the problem perhaps ;-) ) to understand how ActiveX is more dangerous than plug-ins.
While that's true to some extent, there are three common Firefox plugins, all of which have had major security holes: Java, Flash and Adobe PDF. Most people don't need Java or PDF plugins, but Flash is harder to get rid of.
There are about a bazillion ActiveX things and most of them probably have major security holes.
And yet slashdotters continue to bitch about internet explorer
Does any browser other than IE support the 'attack kit of ActiveX exploits' used as the primary vector in this attack?
How did anyone make Linux in the first place if they didn't learn about it in school?
In case you hadn't noticed, Linus went to school in Finland.
Also, haven't Oracle been supporting btrfs development? That may not be doing that much longer if they now own ZFS.
Obviously development would continue without such support as it is GPL and is important for Linux in the future, but perhaps not at the same rate.
By the time you get around to setting up that Solaris server, Btrfs will have stabilized through 3-4 more mainline kernel releases.
Which means that about five years later it will be ready for production use :).
I am thinking of switching one of my non-vital Linux systems to btrfs before long to try it out, but the whole point of setting up a ZFS server would be for proven reliable storage.
Why would a 'weekend hacker' work on OpenSolaris when they could work on Linux instead?
That's the fundamental problem: OpenSolaris has user features that Linux doesn't -- assuming Oracle continue to support it I'm probably going to set up an OpenSolaris server in the next year or so because ZFS is better than anything Linux currently has -- but it doesn't really offer anything to the average 'weekend hacker' that Linux doesn't.
Even if it was made available under the GPL, I suspect most of the best code would be copied into Linux and then it would die off.
I'd say they're much closer in peril-level than the sting/gunshot example above. They both pose a threat to the liberties, freedoms, and lives of Americans.
Lives, perhaps, but exactly what threat does bin Laden pose to Americans' liberties and freedoms? Is he going to run for President in 2012?
All the post-9/11 impositions on American freedoms have come from the US government, not some crazy guy in a cave in Afghanistan. And most of them are things the government have wanted to do for years but had no excuse to impose before that point.
The last thing Linux needs is a set-in-stone kernel interface: 'backwards compatibility' is what has ensured that Windows remains a steaming pile of kludges and security holes as no old components can be thrown away.
I can only presume that you are actually Bill Gates and want to destroy Linux by forcing it to repeat Windows' mistakes.
Once a machine gets owned it's gone. Total wipe, reinstall from good backup. No matter what OS or even WIndows it is.
Joe Sixpack doesn't have a backup.
Also, Joe Sixpack probably don't have XP CDs, so he has to install from the 'recovery partition'; I wonder whether any rootkits are installing themselves into the recovery partition so they'll automatically be reinstalled if someone tries to wipe their system and reinstall from scratch?