The same argument applies for a single disk drive. No one in their right mind uses a single disk drive on data that they care about.
Multiple disk drives increase the chances of disk-related data loss, but failure of a cooling fan does, too. It is incorrect to assume that a two drive RAID 0 is twice as likely to result in data loss as one drive since you need to consider the entire system and the environment it is in.
Now, is RAID-0 is considerably less reliable than a single hard drive? Depends on how you define "considerably". If you have a hot environment with poor airflow and poor power line quality and no UPS, then the answer is no.
Proprietary code makes it illegal for me to help my neighbor? Come on, "proprietary code" isn't law. It doesn't make anything illegal.
Novell, as copyright holder can multi-license without violating the spirit of the GPL. Since they choose to do so, they require contributors be compatible with it in order to accept their contributions. If you believe that this goes against the spirit of the GPL then you recognize the GPL's desire to force it's idealogy on other people's work. That is, of course, fact. Anything that's more proprietary or more permissive than the GPL goes against the spirit of the GPL in that sense. Hell, RMS doesn't even like the LGPL.
No form of greed or self-interest? Surely you must be kidding! Selling a proprietary product is not the only form of greed / self-interest.
RMS and the GPL are certainly driven by self-interest---interest in maximizing the volume of software available to them in source form. RMS doesn't care that you have access to his source. He wants access to yours.
I think you misunderstand what Core Image is for (and what a line of code is for that matter). You won't be building a Photoshop killer out of a "modern foundation for video services".
I don't think system vendors agree with you on the battery life advantage. Fact was that the Crusoe got some design wins early with Japanese manufacturers by buying into those designs. Subsequent generations designed the part out. Frankly, I don't think Transmeta has anything to offer the industry apart from spurring Intel into producing low-power parts.
Curious that you see no requirement for a single CPU to get faster yet desire to see multi-CPU configurations become the norm. That doesn't make any sense. Two half speed CPU's are a liablity compare to a single full speed one though a system designed in that manner may be much easier to build.
The future may be MP via multicore processors because that's the easiest path to superior performance. It won't be because there's no demand for a faster single processor though.
Processor manufacturers have to compete with something and so far it is raw speed. VIA does it on cost, small dies, and low power but their popularity is limited. That's a real market proof that fan-less has limited appeal unless the performance is truly competitive.
More likely they spend the extra power monitoring the fancy "no moving parts" buttons, even when the power is "off", because sizzle sells. Don't kid yourself, the iPod's crappy battery performance has nothing to do with "superior audio performance".
Curious, I knew that about that without having to buy one. Sony did, after all, advertise that aspect of the design very plainly. You are to blame for buying into such a piece of shit.
God would not have designed the iPod the way Apple did. I spoke with Him the other day about it and we agreed that the iPod should have a proper on/off switch so the batteries won't discharge in just a few days. He also doesn't care for how the slipcase scratches the display.
Perhaps we need a player designed by Satan. I bet it'd be way cooler than God's one and you might get sex just for owning one (though not from any mac lovers).
The customer would go to the place they purchased it. That means Real. Did Real promise that music you brought from them would play on all future players for eternity? Unlikely. More likely they promised that it wouldn't (that's what DRM is for, after all). Many seem to be making this point but the fact is that it doesn't matter and neither Apple nor Real actually care about it.
iPod is just a music player. iTunes is a complementary product that not all iPod owners use and is not "bundled" with the iPod. PC manufacturers aren't clamoring to ship iTunes on their PC's now so I wouldn't make too big a claim about MS pressuring them to do otherwise. Windows ships with an MS media player, of course.
I've experienced the elegance of iPod with XPlay and don't prefer the "elegance" of iTunes. Of course Apple wants users to use everything from them since it means all the money goes to them. It wasn't that long ago that Apple insisted you buy a Mac to enjoy the "elegance of the integration". What horseshit was that! XPlay put an end to that.
Apple is not at all worried that they have to support Real. In fact, they hope there are problems so they can say "I told you so." Apple is purely and simply worried about their ability to maintain vendor lockin.
Except that, from the customer's perspective, Real is right and Apple is wrong. Having the choice to use Real's service with the iPod will be nothing but a potential benefit to customers while locking that out only benefits Apple. No doubt both companies want to make money but at least Real isn't trying to screw the customer in order to do so.
Using rsync to back up data. That's the first really interesting interoperability argument I've heard. rsync solution for Windows are amazingly pathetic, though it really shouldn't be hard to change that. All there seems to be are cygwin rsync ports with stupid frontends. I'd like to see a cygwin rsync actually work on my Windows box.
The selling point of the iPod is that it is a seamless integration with iTunes? Hardly. I thought the selling point of the iPod was that it allowed you take a large amount of music with you. iTunes lasted a day on my computer before the abomination was removed. I don't like jukebox software in general but there is better than iTunes and XPlay offers a better connection solution than iTunes does.
Just where do you get the idea that patches will be required on the iPod? Certainly Apple may "require" it when the attempt to break Real's solution.
At least on a PC, special software is required. The iPod does not appear as a FW/USB HDD. It wouldn't matter anyway as the filesystem is foreign and there's the issue of the database. XPlay provides an Explorer plugin that works great and is a far better choice than either iTunes for Windows or the MusicMatch abortion that preceeded it. No way I'd use iTunes for Windows as it is not a better jukebox than existing software, I have no interest in iTMS, and XPlay provides a file manager interface that allows me to control my music directly. No iTunes is a win-win on Windows.
which reminds my of how pathetic I thought the rerelease of the commercial was. What it really said was "We said we would revolutionize the world, destroy the groupthink of the IBM mentality, but what we really meant is that 20 years later we would be hawking nothing more than this little music player you clip to your belt." If you're going to talking trash you need to back it up and Apple sure hasn't lived up to its mouth. I could think of nothing more belittling to Apple's reputation yet they did it themselves.
As in the enemy of my enemy is my friend? They never were the good guys. They've always been closed, proprietary, predatory and litigious. Nothing has changed.
Writing to disk is not an expensive operation from the CPU's perspective, and while software RAID makes it somewhat worse it is not "orders of magnitude".
Disk IO is characterized by enormous waits for access latency followed by (typically) brief busmaster DMA data transfers. Virtually no CPU load exists.
Offloading parity generation from today's processors is meaningless for smaller configurations. Of course, the right answer is to not be doing parity at all. RAID 5 is the work of the devil.
My Mylex controller bit the big one, so I called Mylex up and requested a replacement (as I was entitled). Three weeks later I hadn't received it so I ordered one. One I got the new one I migrated the data off (quite a challenge) and trashed the system. Three weeks later Mylex delivers my replacement. Glad I wasn't depending on them.
Don't depend on a proprietary hardware RAID card when you don't have to and never use Mylex. 3Ware is NOT better. They crash when they lose a drive. Some availability!
Har har! As if that actually helps or the controller manufacturers themselves don't do it with CPU that actually slower than the host. The only thing that's important is the the XOR's occur within one rev of the disks so that you don't drop a rev on your RMW. Plenty of time for that.
Fact is that there's little factual basis for the claim that hardware RAID is better. It all boils down to what solutions are most reliable and my experience is that all the low cost hardware RAID controllers are crap.
Use cheap controllers, software RAID 0, and rsync.
Curious that no one mentions the stability of current x86-64 Linux implementations. I'm running one (SuSE 9.1) and it's very disappointing. Binary software doesn't recognize the processor type and browser plugins don't work. 32 bit browsers would fix that but they are unstable for me. In fact, firefox in any form locks my machine consistently. I certainly wouldn't use the machine for any critical work though most things seem fine.
Any user of a 64 bit x86 system should expect all 32 bit applications for that system to "just work". That's certainly not the case for linux and I expect Microsoft has a much higher standard in that regard.
"If you are Jim, what the hell difference does it make if you are on machine A or B?"
Because there is more than one "Jim". It may not matter to you but it sure as hell matters to the machines and to the network.
There is no way to prevent admins with local and network admin rights from creating accounts with the same name and it makes no sense to assume that such accounts are intended for the same user. Associating the account name with the scope makes perfect sense.
Of course, the UNIX solution is to "hope" that the ID's themselves are consistently assigned since there's only global scope. Talking about "total crap" and "quite annoying".
Slashdot Mirror
The same argument applies for a single disk drive. No one in their right mind uses a single disk drive on data that they care about.
Multiple disk drives increase the chances of disk-related data loss, but failure of a cooling fan does, too. It is incorrect to assume that a two drive RAID 0 is twice as likely to result in data loss as one drive since you need to consider the entire system and the environment it is in.
Now, is RAID-0 is considerably less reliable than a single hard drive? Depends on how you define "considerably". If you have a hot environment with poor airflow and poor power line quality and no UPS, then the answer is no.
Proprietary code makes it illegal for me to help my neighbor? Come on, "proprietary code" isn't law. It doesn't make anything illegal.
Novell, as copyright holder can multi-license without violating the spirit of the GPL. Since they choose to do so, they require contributors be compatible with it in order to accept their contributions. If you believe that this goes against the spirit of the GPL then you recognize the GPL's desire to force it's idealogy on other people's work. That is, of course, fact. Anything that's more proprietary or more permissive than the GPL goes against the spirit of the GPL in that sense. Hell, RMS doesn't even like the LGPL.
No form of greed or self-interest? Surely you must be kidding! Selling a proprietary product is not the only form of greed / self-interest.
RMS and the GPL are certainly driven by self-interest---interest in maximizing the volume of software available to them in source form. RMS doesn't care that you have access to his source. He wants access to yours.
I think you misunderstand what Core Image is for (and what a line of code is for that matter). You won't be building a Photoshop killer out of a "modern foundation for video services".
I don't think system vendors agree with you on the battery life advantage. Fact was that the Crusoe got some design wins early with Japanese manufacturers by buying into those designs. Subsequent generations designed the part out. Frankly, I don't think Transmeta has anything to offer the industry apart from spurring Intel into producing low-power parts.
Curious that you see no requirement for a single CPU to get faster yet desire to see multi-CPU configurations become the norm. That doesn't make any sense. Two half speed CPU's are a liablity compare to a single full speed one though a system designed in that manner may be much easier to build.
The future may be MP via multicore processors because that's the easiest path to superior performance. It won't be because there's no demand for a faster single processor though.
Processor manufacturers have to compete with something and so far it is raw speed. VIA does it on cost, small dies, and low power but their popularity is limited. That's a real market proof that fan-less has limited appeal unless the performance is truly competitive.
"I've opened up a number of Dell boxes only to find below-spec power supplies and such."
Right. Identified by a "below-spec" sticker on them I assume. What's the "and such" specifically?
Using Pentium-M as a desktop processor won't take several years.
More likely they spend the extra power monitoring the fancy "no moving parts" buttons, even when the power is "off", because sizzle sells. Don't kid yourself, the iPod's crappy battery performance has nothing to do with "superior audio performance".
Curious, I knew that about that without having to buy one. Sony did, after all, advertise that aspect of the design very plainly. You are to blame for buying into such a piece of shit.
God would not have designed the iPod the way Apple did. I spoke with Him the other day about it and we agreed that the iPod should have a proper on/off switch so the batteries won't discharge in just a few days. He also doesn't care for how the slipcase scratches the display.
Perhaps we need a player designed by Satan. I bet it'd be way cooler than God's one and you might get sex just for owning one (though not from any mac lovers).
The customer would go to the place they purchased it. That means Real. Did Real promise that music you brought from them would play on all future players for eternity? Unlikely. More likely they promised that it wouldn't (that's what DRM is for, after all). Many seem to be making this point but the fact is that it doesn't matter and neither Apple nor Real actually care about it.
iPod is just a music player. iTunes is a complementary product that not all iPod owners use and is not "bundled" with the iPod. PC manufacturers aren't clamoring to ship iTunes on their PC's now so I wouldn't make too big a claim about MS pressuring them to do otherwise. Windows ships with an MS media player, of course.
I've experienced the elegance of iPod with XPlay and don't prefer the "elegance" of iTunes. Of course Apple wants users to use everything from them since it means all the money goes to them. It wasn't that long ago that Apple insisted you buy a Mac to enjoy the "elegance of the integration". What horseshit was that! XPlay put an end to that.
Apple is not at all worried that they have to support Real. In fact, they hope there are problems so they can say "I told you so." Apple is purely and simply worried about their ability to maintain vendor lockin.
Except that, from the customer's perspective, Real is right and Apple is wrong. Having the choice to use Real's service with the iPod will be nothing but a potential benefit to customers while locking that out only benefits Apple. No doubt both companies want to make money but at least Real isn't trying to screw the customer in order to do so.
Using rsync to back up data. That's the first really interesting interoperability argument I've heard. rsync solution for Windows are amazingly pathetic, though it really shouldn't be hard to change that. All there seems to be are cygwin rsync ports with stupid frontends. I'd like to see a cygwin rsync actually work on my Windows box.
The selling point of the iPod is that it is a seamless integration with iTunes? Hardly. I thought the selling point of the iPod was that it allowed you take a large amount of music with you. iTunes lasted a day on my computer before the abomination was removed. I don't like jukebox software in general but there is better than iTunes and XPlay offers a better connection solution than iTunes does.
Just where do you get the idea that patches will be required on the iPod? Certainly Apple may "require" it when the attempt to break Real's solution.
Change your threshold and you will. Nothing gets modded down faster than an anti-iPod post.
At least on a PC, special software is required. The iPod does not appear as a FW/USB HDD. It wouldn't matter anyway as the filesystem is foreign and there's the issue of the database. XPlay provides an Explorer plugin that works great and is a far better choice than either iTunes for Windows or the MusicMatch abortion that preceeded it. No way I'd use iTunes for Windows as it is not a better jukebox than existing software, I have no interest in iTMS, and XPlay provides a file manager interface that allows me to control my music directly. No iTunes is a win-win on Windows.
which reminds my of how pathetic I thought the rerelease of the commercial was. What it really said was "We said we would revolutionize the world, destroy the groupthink of the IBM mentality, but what we really meant is that 20 years later we would be hawking nothing more than this little music player you clip to your belt." If you're going to talking trash you need to back it up and Apple sure hasn't lived up to its mouth. I could think of nothing more belittling to Apple's reputation yet they did it themselves.
As in the enemy of my enemy is my friend? They never were the good guys. They've always been closed, proprietary, predatory and litigious. Nothing has changed.
Writing to disk is not an expensive operation from the CPU's perspective, and while software RAID makes it somewhat worse it is not "orders of magnitude".
Disk IO is characterized by enormous waits for access latency followed by (typically) brief busmaster DMA data transfers. Virtually no CPU load exists.
Offloading parity generation from today's processors is meaningless for smaller configurations. Of course, the right answer is to not be doing parity at all. RAID 5 is the work of the devil.
My Mylex controller bit the big one, so I called Mylex up and requested a replacement (as I was entitled). Three weeks later I hadn't received it so I ordered one. One I got the new one I migrated the data off (quite a challenge) and trashed the system. Three weeks later Mylex delivers my replacement. Glad I wasn't depending on them.
Don't depend on a proprietary hardware RAID card when you don't have to and never use Mylex. 3Ware is NOT better. They crash when they lose a drive. Some availability!
Har har! As if that actually helps or the controller manufacturers themselves don't do it with CPU that actually slower than the host. The only thing that's important is the the XOR's occur within one rev of the disks so that you don't drop a rev on your RMW. Plenty of time for that.
Fact is that there's little factual basis for the claim that hardware RAID is better. It all boils down to what solutions are most reliable and my experience is that all the low cost hardware RAID controllers are crap.
Use cheap controllers, software RAID 0, and rsync.
You won't with RAID either. Rsync can be run without the -delete option to avoid file deletions on you backup.
Curious that no one mentions the stability of current x86-64 Linux implementations. I'm running one (SuSE 9.1) and it's very disappointing. Binary software doesn't recognize the processor type and browser plugins don't work. 32 bit browsers would fix that but they are unstable for me. In fact, firefox in any form locks my machine consistently. I certainly wouldn't use the machine for any critical work though most things seem fine.
Any user of a 64 bit x86 system should expect all 32 bit applications for that system to "just work". That's certainly not the case for linux and I expect Microsoft has a much higher standard in that regard.
"If you are Jim, what the hell difference does it make if you are on machine A or B?"
Because there is more than one "Jim". It may not matter to you but it sure as hell matters to the machines and to the network.
There is no way to prevent admins with local and network admin rights from creating accounts with the same name and it makes no sense to assume that such accounts are intended for the same user. Associating the account name with the scope makes perfect sense.
Of course, the UNIX solution is to "hope" that the ID's themselves are consistently assigned since there's only global scope. Talking about "total crap" and "quite annoying".