How wise is it to prevent those who owe you money from getting jobs to payback said money?
How wise is it to not exercise the only collateral you have to get paid on what you are owed?
This is like someone complaining they can't sell their car because they don't have the title on it, because they refuse to pay off the loan. Imagine that, it's a great incentive to pay off the loan.
The borrower may claim that preventing them from selling the car is hampering their ability to get cash to help pay off the loan. Soooo, does that make it a wise idea for the bank to hand them the title in the hopes that someday the borrower actually pays off the loan? That's just all kinds of stupid. Collateral exists for a reason, and that's all the transcripts are - collateral on a debt.\
That doesn't mean the whole idea of digging yourself a decade into debt with college loans to get a degree to get a job is a good plan. But if you choose to go that route, to get a loan through the college instead of say, through your bank or relatives, you accept that they will hold the transcripts if you are in default. You know this going into the deal. I'm not having any pity for you if later you think it's not fair. It's the road you chose, and it's not the only road that was available to you. You thought it was fair when you took out the loans. Now take responsibility for your decisions and stop blaming others for taking advantage of your bad choices and bad fortune.
If you think about it, this could provide more information on your opponents. Though it is a bit of a gamble - can you get valuable information without too much risk? Or, is it worth the risk?
Think about the whole process of infiltration. Once you get your foot in the door you start gathering information and testing the waters to see what you can do. If you don't think you've been discovered, but you have, then the defenders have some good opportunities. They can feed you false intelligence, make you think you are burrowing into an important control system that's actually a honeypot, give them a false sense of accomplishing their goal, waste their time and resources. Done properly, this is very useful counter-intelligence.
Fooling the other guy is valuable. Tricking the other guy into thinking he's fooled you can be even more valuable. I think that's the core of what this is about. But as I said before, it's a risk, and could get out of control.
BASIC raid5 requires identical capacity drives. Intelligent raid5 does not. Imagine a four drive array, containing three 40gb drives and one 30gb drive. So you can treat the first 30gb of the 40's, and the 30gb as one segment of the storage, (3 data and a parity, total of 90gb of protected storage) and the remaining 10gb of space on the three 40's run as 2 data with a parity, for an additional 20gb of protected storage. (if you tried to just ignore the oversize on the 40's you'd lose out on almost 20% of your potential capacity) Those sizes aren't realistic but you can see the application easier.
You just organize the data groups to each span as many drives as possible, to make the most efficient use of the storage. It's not a requirement to use the same grouping strategy on all groups within the raid. Some of the groups may even end up as mirrors. (as would be the case with two 40's and a 30, the extra 10 on the 40's would be mirrored to recover an additional 10gb of space)
At each stage, the agency has to decide to invest every more resources looking deeper (they have finite resources). Having good layers of steganography is more likely to make them spend resources elsewhere than having worse layers.
While layering steno is certainly possible and may be a good idea in theory, I've never seen it applied. I think once you've already raised the eyebrow through the roof by layer1 stego'ing, they're going to get out a fine tooth comb on whatever they got out of that layer. If they're even halfway professional they'd be considering a layer2 steno, and you're already under the microscope at that point. I doubt it would help at all. Steno's purpose is to avoid suspicion and being scrutinized, and if you're already past layer 1 that's shot to hell.
Can be done very simply with a little bit of tcpdump. they're just sniffing network traffic from the machine to reveal information that skype doesn't normally display.
They make it sound like it's some awesome service hacking app when it's just displaying information the client app is just choosing not to show you.
Does this really surprise anyone? Skype directly connects you to another user. Their servers aren't a relay, they're just a meeting point to hook users together. Both users' computers simply have to have the IP address of the other person for their service to function. (though I could see them relaying just text traffic, but certainly not audio/video)
Steno doesn't require encryption. It prevents the information from getting into the hands of the opposition to begin with. They are probably under the assumption that trying to encrypt something (A) is a waste of time if they are likely to have the resources to break it and (B) will attract attention.
Analysis of randomness is quite an advanced science. It's not nearly as difficult as you might imagine to spot an anomaly in random data. Few things in life are truly random, and if you are familiar enough with the kind of randomness in something, you will have a very good statistical chance of noticing it.
As for the steno itself, there's a double-edged sword there. The same as encryption, only a fool designs his own. Without a really deep understanding of encryption it's easy to make a subtle mistake in you design that introduces a significant weakness. So on that hand we have to assume they are using something at least somewhat commonly available. But that's where the other edge gets you... it becomes MUCH easier to spot steno'd data when you are checking for a list of specific methods of steno. The analysis tool will scan the data against dozens or hundreds of common tools, and draw a nice graph with the line mostly hanging around the 2% point at the bottom, except for one eyesore of a spike for one of the tests. Then they take off the gloves and get to work.
Just tossing out a very basic example, lets say they are steno'ing encrypted data into a big tiff by overwriting the LSB on teach byte. Visually you won't notice this because the difference is too small Mathematically the data you're storing is more-or-less flat random. BUT the data you replaced ISN'T. The LSB in an image is far from noise, and will have areas that are related by their relative position in the image. This will stick out like a sore thumb when you run a little analysis on the bits in the image. It'll be too random. And at this point your steno is busted, and it comes down to breaking the encryption. The lesson there is pretty basic - encrypted data will stick out about as well as data in the clear. All you're doing at that point is buying a little time. And intel agencies are both persistent and patient. You're better off investing more time in better steno.
To throw in an analogy, look at smuggling. If a truck is being searched at the border and the smugglers did a good job distracting from the small hide where the goods are, they get past the checkpoint with zero problem. It doesn't matter if the goods are in a safe in the hole or just tossed in a grocery sac, being in a safe didn't help. On the other hand, if the guards notice the missing volume, you're done. It doesn't matter how well you've concealed the opening, it doesn't matter how sturdy the latch, it doesn't matter if the goods are in a safe. You've been found out, and you're done. They will tear the truck apart until they get into the hole, and bust open the safe. The same principles are at work with steno.
It certainly would be nice for some of these billion dollar infringement judgements to go to the music creators. But they probably stood to lose a few dollars. (assuming they ever got out of the hole with the record label they signed on with) Shows just how screwed up the economics of this issue is, on both ends of the line.
I'd wager that number of agents that could be bribed to allow explosives/weapons/etc on board is significantly smaller than those that could be bribed to allow drugs.
I'd also wager that said screener was bribed t let the smuggler through with a 10 lb brick of whatever the heck he was carrying. I don't think the screener pulled out the DEA drug test kit to make sure that was a block of coke and not a block of say, semtex. It's probably also fair to assume that this wasn't the first time he's done this, this is just the first time getting caught at it. After the first few, the screener that was getting used to the quick cash probably was paying a lot less attention to what the courier was carrying.
The only reason it got caught this time is the smuggler screwed up. That indicates there's a fundamental flaw in the system. But it doesn't really matter, nobody that can fix it is listening to anything we have to say.
the problem here is that the US is *known* to be storing ALL email traffic that routes through the united states. Sounds like a daunting task, but there's a reason they have all these big high security data centers all over the place and have "high security rooms" at all the telcos and large ISPs. That traffic gets siphoned off to their data centers for storage for later in case they need it. There's a simple reason why those places have petabytes of storage.
So there is never a question of "but they'd have to have been watching for that email last week/month/year and it's long since been sent and removed from caches". No. They have it. They have them all, just in case. Watch Enemy of the State. Watch how they pull up satellite footage from hours and days ago. Same principle here, if you can record everything, it works like a time machine. (for the past anyway)
So yes, busting down a door and taking the remailer keys gives them 100% access to 100% of the traffic that has been sent by that remailer at ANY point in the past where it crossed through a US ISP.
The truly disgusting part of this is they got the KEYS. Technically all they NEEDED was to hand over the encrypted message to the AU authorities, they break down the door and use the key to decode the message, and turn over the message, then wipe their copy of the key. That would be the "proper" way to do it, not to abuse the system, but instead they handed over the KEYS themselves, and now the US can decrypt truckloads of hard drives of emails that they have NO business having access to. That is the true crime here. It's like having a legal reason to subpoena a safe deposit box at a bank, and the bank hands them over a master key that opens every box in the vault and lets them look through anything they want. That's just WRONG.
Every time someone sends a bomb threat they can pull this stunt, it's like christmas over at the NSA, "we got another key! lets see what goodies we can find!" Talk about an incentive for abuse... Normally I don't go "tinfoil hat" on things, but THIS is actually an instance where I could start to buy into someone suggesting the NSA/etc forging a bomb threat just to get access to another random footlocker of encrypted data they want a peek at.
If we dont let them send bomb threats, we're undermining free speech and the Internet"
To which I reply "They need to find a different way to discourage or stop them from sending bomb threats. Inflicting me with collateral damage in the quest for better law enforcement is unacceptable, and so is removing my ability to speak with anonymity."
Given the choice, I think I'd rather deal with the occasional bomb threat than not be able to speak anonymously.
All games with in-game purchases are designed to entice the user to buy the in-game perks. Age doesn't matter. All that matters is who has control of the credit card. A parent that gives their kid access to their card is responsible for how they use it, whether they like it or not.
Parent gives kid a BB-gun, kid shoots their dog in the eye, parent tries to blame Daisy. NO
Quit trying to make me your kid's guardian. Accept your responsibility.
Sounds like another good example of people wanting the world to do their parenting for them. "My kid's doing something I don't like, and you're helping them do it! This is all your fault!" No, we're not stopping them, and neither are you. Keeping a handle on what your kids do is your responsibility, not mine. Start being a more responsible parent.
You gave them the password to your cc-bound account, you didn't effectively train them in what is acceptable and what is not, and you didn't keep track of their purchases. You failed start, middle, and end. You denied your responsibility, expecting "someone else" to take care of your child for you, and now you are responsible for the outcome. I'm glad to see this go to court. This way it will (assuming sanity prevails) establish better precedent for this sort of irresponsibility we read about from time to time.
"wouldn't boot" is pretty general... Apple's Repair Extension Programs cover specific models, for a limited additional duration, that are demonstrating specific symptoms which are being caused by a very specific problem.
We get people in from time to time with a computer that has an REP on that model, and they're expecting Apple to cover something else out of warranty. And we get people demanding we "fix" the computer simply because there is an REP on that model even though it's working fine. That's not how REPs work. (we had a school show up with a panel van FULL of emacs that were listed on an REP, we tested ALL of them and repaired TWO)
If it powered on, and chimed (possibly after clearing pram to turn the volume back up) but showed no video, and could be heard to boot up (hdd access) and possibly even get interaction from it (turning volume up and down and hearing the reply) and external video was also dead, THAT should be covered and I would be surprised if they didn't cover it. All AASPs were given a special tool to test the computer and verify the problem also, and this test could be run after the video was out, OR before it was out, and could identify a computer that was beginning to fail, even if the symptoms were very minor or infrequent. If it was demonstrating symptoms, this test should have been run on your computer. Users tend to put off taking computers in for service, so only about 3% of the machines I saw with this problem still had usable video by the time they checked them in.
However, if it failed to turn on, or failed to post at all, no, that's not the issue the REP was for, that's not a video problem, and you were not entitled to coverage by this REP. I also ran into a couple that were having "no video" problems but that didn't fail the special test app, those users had to pay for their repair because it was the same part, but not the same cause.
I won't say that every AASP and Apple Store does the right thing. I'm just sayin' how it's supposed to work, and how I make it work here. If you still think they didn't react correctly, call applecare (even now) and talk with them about it. I've seen them make things right for people outwards of 5 years after purchase, with a discount on a new machine for example.
No, they were footing the bill for their own screw-up. The part may have been defective, and nVidia may have made it, but Apple approved it for use in their own products based on, clearly, insufficient validation that the part met their standards. In the real world, that's how it works.
Check your facts. (or start using them) This technically wasn't even a manufacturing defect. It was a specifications defect. Nvidia provides manufacturers with very detailed information on their raw products, including cooling requirements. They understated the cooling requirements for the chips. As a result, Apple and several other manufacturers didn't design in sufficient cooling to keep the chips safely cooled under long duration, high stress conditions. Apple and the other manufacturers bore very limited fault in the problem. (it's impractical to hold off on releasing a new technology product for 2 months while you stress test it enough to catch something like this) So when users started playing WoW or something else GPU-intensive for long sessions, it would stress and over time eventually crack the traces on the package.
Nvidia quietly reworked their manufacturing process to make the part match the specs, but the damage was done, those parts were already installed and in users' hands.
If I'm a cabinet manufacturer and I order hinges that are rated at 100,000 operations I expect them to last that long on the average, you can't expect me to make some cabinets and then open and close them 50,000 times to make sure the hardware is as durable as the manufacturer claims it is before selling any. Manufactures have to put some trust in their OEMs' specs.
If Apple promised to replace parts they knew to be defective for up to four years then they'd better step up. The court costs far outweighed the costs of replacing the GPU in his MacBook (and probably 100's of others). The cost of the bad press? It seems Apple has always been willing to test the patience of their customers' loyalties.
For more details on the problem check this link. Here's the header:
In July 2008, NVIDIA publicly acknowledged a higher than normal failure rate for some of their graphics processors due to a packaging defect. At that same time, NVIDIA assured Apple that Mac computers with these graphics processors were not affected. However, after an Apple-led investigation, Apple has determined that some MacBook Pro computers with the NVIDIA GeForce 8600M GT graphics processor may be affected. If the NVIDIA graphics processor in your MacBook Pro has failed, or fails within four years of the original date of purchase, a repair will be done free of charge, even if your MacBook Pro is out of warranty.
I'm surprised anyone has been refused replacement inside 4 years. I bet I've repaired around 170 of these units for this problem, and I have only just recently started seeing Apple refuse a warranty repair, because the computers are starting to cross 4 years old. The only time I see problems of this nature is if they purchase old stock and don't register. Apple assumes a computer is sold 30 days after manufacture if you don't send in your registration. If you need warranty service and are on the edge you may need to submit your proof of purchase to update your purchase date on record with Apple to get warranty coverage. Maybe that played into this case?
And this problem stems not from Apple, but from Nvidia. I started seeing this issue on new machines a few months after this model was first released, and Apple started going rounds with Nvidia around the 10 month mark, just before these machines were going to start falling out of the 1 year warranty. Nvidia insisted this was not a defect and refused to cover anything. We had to start refusing repairs for some machines after the 1 year mark. Then about 2 months after that I found that Apple had gotten sick of Nvidia stalling and denying, and decided to cover these repairs, before they had even gotten Nvidia to budge. Apple sent notice to users that had paid for a repair that would now be considered covered, and refunds were issued. Apple started the repair extension program for this issue and covered repairs from that point forward. This was months before Nvidia was forced to accept responsibility and start reimbursing Apple for the defect.
So I find it unfortunate that Apple is receiving a lot of FUD and bad press on this. They do tend to go the extra mile for their customers, they're consistently rated at the TOP for customer service. They were footing the bill for Nvidia's screw-up long before they were guaranteed of getting anything back. Try to find an example of that from any of the other computer manufacturers out there.
Slashdot Mirror
How wise is it to not exercise the only collateral you have to get paid on what you are owed?
This is like someone complaining they can't sell their car because they don't have the title on it, because they refuse to pay off the loan. Imagine that, it's a great incentive to pay off the loan.
The borrower may claim that preventing them from selling the car is hampering their ability to get cash to help pay off the loan. Soooo, does that make it a wise idea for the bank to hand them the title in the hopes that someday the borrower actually pays off the loan? That's just all kinds of stupid. Collateral exists for a reason, and that's all the transcripts are - collateral on a debt.\
That doesn't mean the whole idea of digging yourself a decade into debt with college loans to get a degree to get a job is a good plan. But if you choose to go that route, to get a loan through the college instead of say, through your bank or relatives, you accept that they will hold the transcripts if you are in default. You know this going into the deal. I'm not having any pity for you if later you think it's not fair. It's the road you chose, and it's not the only road that was available to you. You thought it was fair when you took out the loans. Now take responsibility for your decisions and stop blaming others for taking advantage of your bad choices and bad fortune.
If you think about it, this could provide more information on your opponents. Though it is a bit of a gamble - can you get valuable information without too much risk? Or, is it worth the risk?
Think about the whole process of infiltration. Once you get your foot in the door you start gathering information and testing the waters to see what you can do. If you don't think you've been discovered, but you have, then the defenders have some good opportunities. They can feed you false intelligence, make you think you are burrowing into an important control system that's actually a honeypot, give them a false sense of accomplishing their goal, waste their time and resources. Done properly, this is very useful counter-intelligence.
Fooling the other guy is valuable. Tricking the other guy into thinking he's fooled you can be even more valuable. I think that's the core of what this is about. But as I said before, it's a risk, and could get out of control.
Worse than that, she probably violated facebook's ToS. Somebody call the cops!
lets just hope that Sony isn't supplying the batteries...
BASIC raid5 requires identical capacity drives. Intelligent raid5 does not. Imagine a four drive array, containing three 40gb drives and one 30gb drive. So you can treat the first 30gb of the 40's, and the 30gb as one segment of the storage, (3 data and a parity, total of 90gb of protected storage) and the remaining 10gb of space on the three 40's run as 2 data with a parity, for an additional 20gb of protected storage. (if you tried to just ignore the oversize on the 40's you'd lose out on almost 20% of your potential capacity) Those sizes aren't realistic but you can see the application easier.
You just organize the data groups to each span as many drives as possible, to make the most efficient use of the storage. It's not a requirement to use the same grouping strategy on all groups within the raid. Some of the groups may even end up as mirrors. (as would be the case with two 40's and a 30, the extra 10 on the 40's would be mirrored to recover an additional 10gb of space)
the meeting was only a secret before he showed up. After that well, there he is. This is hardly a leak OR a secret, more like "breaking news".
It would be a leak/secret if it came up before he landed, granted.
While layering steno is certainly possible and may be a good idea in theory, I've never seen it applied. I think once you've already raised the eyebrow through the roof by layer1 stego'ing, they're going to get out a fine tooth comb on whatever they got out of that layer. If they're even halfway professional they'd be considering a layer2 steno, and you're already under the microscope at that point. I doubt it would help at all. Steno's purpose is to avoid suspicion and being scrutinized, and if you're already past layer 1 that's shot to hell.
Can be done very simply with a little bit of tcpdump. they're just sniffing network traffic from the machine to reveal information that skype doesn't normally display.
They make it sound like it's some awesome service hacking app when it's just displaying information the client app is just choosing not to show you.
Does this really surprise anyone? Skype directly connects you to another user. Their servers aren't a relay, they're just a meeting point to hook users together. Both users' computers simply have to have the IP address of the other person for their service to function. (though I could see them relaying just text traffic, but certainly not audio/video)
Steno doesn't require encryption. It prevents the information from getting into the hands of the opposition to begin with. They are probably under the assumption that trying to encrypt something (A) is a waste of time if they are likely to have the resources to break it and (B) will attract attention.
Analysis of randomness is quite an advanced science. It's not nearly as difficult as you might imagine to spot an anomaly in random data. Few things in life are truly random, and if you are familiar enough with the kind of randomness in something, you will have a very good statistical chance of noticing it.
As for the steno itself, there's a double-edged sword there. The same as encryption, only a fool designs his own. Without a really deep understanding of encryption it's easy to make a subtle mistake in you design that introduces a significant weakness. So on that hand we have to assume they are using something at least somewhat commonly available. But that's where the other edge gets you... it becomes MUCH easier to spot steno'd data when you are checking for a list of specific methods of steno. The analysis tool will scan the data against dozens or hundreds of common tools, and draw a nice graph with the line mostly hanging around the 2% point at the bottom, except for one eyesore of a spike for one of the tests. Then they take off the gloves and get to work.
Just tossing out a very basic example, lets say they are steno'ing encrypted data into a big tiff by overwriting the LSB on teach byte. Visually you won't notice this because the difference is too small Mathematically the data you're storing is more-or-less flat random. BUT the data you replaced ISN'T. The LSB in an image is far from noise, and will have areas that are related by their relative position in the image. This will stick out like a sore thumb when you run a little analysis on the bits in the image. It'll be too random. And at this point your steno is busted, and it comes down to breaking the encryption. The lesson there is pretty basic - encrypted data will stick out about as well as data in the clear. All you're doing at that point is buying a little time. And intel agencies are both persistent and patient. You're better off investing more time in better steno.
To throw in an analogy, look at smuggling. If a truck is being searched at the border and the smugglers did a good job distracting from the small hide where the goods are, they get past the checkpoint with zero problem. It doesn't matter if the goods are in a safe in the hole or just tossed in a grocery sac, being in a safe didn't help. On the other hand, if the guards notice the missing volume, you're done. It doesn't matter how well you've concealed the opening, it doesn't matter how sturdy the latch, it doesn't matter if the goods are in a safe. You've been found out, and you're done. They will tear the truck apart until they get into the hole, and bust open the safe. The same principles are at work with steno.
It certainly would be nice for some of these billion dollar infringement judgements to go to the music creators. But they probably stood to lose a few dollars. (assuming they ever got out of the hole with the record label they signed on with) Shows just how screwed up the economics of this issue is, on both ends of the line.
I'd also wager that said screener was bribed t let the smuggler through with a 10 lb brick of whatever the heck he was carrying. I don't think the screener pulled out the DEA drug test kit to make sure that was a block of coke and not a block of say, semtex. It's probably also fair to assume that this wasn't the first time he's done this, this is just the first time getting caught at it. After the first few, the screener that was getting used to the quick cash probably was paying a lot less attention to what the courier was carrying.
The only reason it got caught this time is the smuggler screwed up. That indicates there's a fundamental flaw in the system. But it doesn't really matter, nobody that can fix it is listening to anything we have to say.
that keep sponsoring these bills?
Time for Alucard to open a can...
the problem here is that the US is *known* to be storing ALL email traffic that routes through the united states. Sounds like a daunting task, but there's a reason they have all these big high security data centers all over the place and have "high security rooms" at all the telcos and large ISPs. That traffic gets siphoned off to their data centers for storage for later in case they need it. There's a simple reason why those places have petabytes of storage.
So there is never a question of "but they'd have to have been watching for that email last week/month/year and it's long since been sent and removed from caches". No. They have it. They have them all, just in case. Watch Enemy of the State. Watch how they pull up satellite footage from hours and days ago. Same principle here, if you can record everything, it works like a time machine. (for the past anyway)
So yes, busting down a door and taking the remailer keys gives them 100% access to 100% of the traffic that has been sent by that remailer at ANY point in the past where it crossed through a US ISP.
The truly disgusting part of this is they got the KEYS. Technically all they NEEDED was to hand over the encrypted message to the AU authorities, they break down the door and use the key to decode the message, and turn over the message, then wipe their copy of the key. That would be the "proper" way to do it, not to abuse the system, but instead they handed over the KEYS themselves, and now the US can decrypt truckloads of hard drives of emails that they have NO business having access to. That is the true crime here. It's like having a legal reason to subpoena a safe deposit box at a bank, and the bank hands them over a master key that opens every box in the vault and lets them look through anything they want. That's just WRONG.
Every time someone sends a bomb threat they can pull this stunt, it's like christmas over at the NSA, "we got another key! lets see what goodies we can find!" Talk about an incentive for abuse... Normally I don't go "tinfoil hat" on things, but THIS is actually an instance where I could start to buy into someone suggesting the NSA/etc forging a bomb threat just to get access to another random footlocker of encrypted data they want a peek at.
Am I arrogant for saying "wasn't this obvious?" Isn't that newton's first law at work? This looks like a job for Captain Obvious.
[grammarnazi] I don't think you can do those two things in that order....[/grammarnazi]
To which I reply "They need to find a different way to discourage or stop them from sending bomb threats. Inflicting me with collateral damage in the quest for better law enforcement is unacceptable, and so is removing my ability to speak with anonymity."
Given the choice, I think I'd rather deal with the occasional bomb threat than not be able to speak anonymously.
All games with in-game purchases are designed to entice the user to buy the in-game perks. Age doesn't matter. All that matters is who has control of the credit card. A parent that gives their kid access to their card is responsible for how they use it, whether they like it or not.
Parent gives kid a BB-gun, kid shoots their dog in the eye, parent tries to blame Daisy. NO
Quit trying to make me your kid's guardian. Accept your responsibility.
Sounds like another good example of people wanting the world to do their parenting for them. "My kid's doing something I don't like, and you're helping them do it! This is all your fault!" No, we're not stopping them, and neither are you. Keeping a handle on what your kids do is your responsibility, not mine. Start being a more responsible parent.
You gave them the password to your cc-bound account, you didn't effectively train them in what is acceptable and what is not, and you didn't keep track of their purchases. You failed start, middle, and end. You denied your responsibility, expecting "someone else" to take care of your child for you, and now you are responsible for the outcome. I'm glad to see this go to court. This way it will (assuming sanity prevails) establish better precedent for this sort of irresponsibility we read about from time to time.
to piss off someone that can make you disappear
"wouldn't boot" is pretty general... Apple's Repair Extension Programs cover specific models, for a limited additional duration, that are demonstrating specific symptoms which are being caused by a very specific problem.
We get people in from time to time with a computer that has an REP on that model, and they're expecting Apple to cover something else out of warranty. And we get people demanding we "fix" the computer simply because there is an REP on that model even though it's working fine. That's not how REPs work. (we had a school show up with a panel van FULL of emacs that were listed on an REP, we tested ALL of them and repaired TWO)
If it powered on, and chimed (possibly after clearing pram to turn the volume back up) but showed no video, and could be heard to boot up (hdd access) and possibly even get interaction from it (turning volume up and down and hearing the reply) and external video was also dead, THAT should be covered and I would be surprised if they didn't cover it. All AASPs were given a special tool to test the computer and verify the problem also, and this test could be run after the video was out, OR before it was out, and could identify a computer that was beginning to fail, even if the symptoms were very minor or infrequent. If it was demonstrating symptoms, this test should have been run on your computer. Users tend to put off taking computers in for service, so only about 3% of the machines I saw with this problem still had usable video by the time they checked them in.
However, if it failed to turn on, or failed to post at all, no, that's not the issue the REP was for, that's not a video problem, and you were not entitled to coverage by this REP. I also ran into a couple that were having "no video" problems but that didn't fail the special test app, those users had to pay for their repair because it was the same part, but not the same cause.
I won't say that every AASP and Apple Store does the right thing. I'm just sayin' how it's supposed to work, and how I make it work here. If you still think they didn't react correctly, call applecare (even now) and talk with them about it. I've seen them make things right for people outwards of 5 years after purchase, with a discount on a new machine for example.
Is that all there are in the USA? I would have expected that to be a much larger number.
(I think I'll forego signing off with my call, cut short one more link in a few three-letter's relational databases)
Check your facts. (or start using them) This technically wasn't even a manufacturing defect. It was a specifications defect. Nvidia provides manufacturers with very detailed information on their raw products, including cooling requirements. They understated the cooling requirements for the chips. As a result, Apple and several other manufacturers didn't design in sufficient cooling to keep the chips safely cooled under long duration, high stress conditions. Apple and the other manufacturers bore very limited fault in the problem. (it's impractical to hold off on releasing a new technology product for 2 months while you stress test it enough to catch something like this) So when users started playing WoW or something else GPU-intensive for long sessions, it would stress and over time eventually crack the traces on the package.
Nvidia quietly reworked their manufacturing process to make the part match the specs, but the damage was done, those parts were already installed and in users' hands.
If I'm a cabinet manufacturer and I order hinges that are rated at 100,000 operations I expect them to last that long on the average, you can't expect me to make some cabinets and then open and close them 50,000 times to make sure the hardware is as durable as the manufacturer claims it is before selling any. Manufactures have to put some trust in their OEMs' specs.
This didn't sound like fire-fighting? Though I'll agree, someone that's always busy fighting fires needs to focus more on fire prevention.
For more details on the problem check this link. Here's the header:
I'm surprised anyone has been refused replacement inside 4 years. I bet I've repaired around 170 of these units for this problem, and I have only just recently started seeing Apple refuse a warranty repair, because the computers are starting to cross 4 years old. The only time I see problems of this nature is if they purchase old stock and don't register. Apple assumes a computer is sold 30 days after manufacture if you don't send in your registration. If you need warranty service and are on the edge you may need to submit your proof of purchase to update your purchase date on record with Apple to get warranty coverage. Maybe that played into this case?
And this problem stems not from Apple, but from Nvidia. I started seeing this issue on new machines a few months after this model was first released, and Apple started going rounds with Nvidia around the 10 month mark, just before these machines were going to start falling out of the 1 year warranty. Nvidia insisted this was not a defect and refused to cover anything. We had to start refusing repairs for some machines after the 1 year mark. Then about 2 months after that I found that Apple had gotten sick of Nvidia stalling and denying, and decided to cover these repairs, before they had even gotten Nvidia to budge. Apple sent notice to users that had paid for a repair that would now be considered covered, and refunds were issued. Apple started the repair extension program for this issue and covered repairs from that point forward. This was months before Nvidia was forced to accept responsibility and start reimbursing Apple for the defect.
So I find it unfortunate that Apple is receiving a lot of FUD and bad press on this. They do tend to go the extra mile for their customers, they're consistently rated at the TOP for customer service. They were footing the bill for Nvidia's screw-up long before they were guaranteed of getting anything back. Try to find an example of that from any of the other computer manufacturers out there.