A 1U box with 128 such slots should be quite doable. A 3.5 inch hard drive sized box with 24 such slots should be quite doable. SATA2 would probably be the ideal interface.
There are really only three pieces that you need. First you need the main controller that talks SATA2 out one end and talks to the flash controllers on the other end. Then you need the interface hardware to connect to so many SD slots. Then you need the software to do the wear leveling and drive emulation.
Notebook size is a bit trickier. A full-sized SD card is 24mm by 32mm and is 2.1mm thick. A typical notebook drive (2.5 inch) is 100mm by 70mm and 9.5mm thick. You will need some space for the housing and interface electronics and you can't stand the SD cards up (assuming you use full-sized SD cards). One possibility would be to have SD slots on both sides of two long narrow vertical boards that run down the center of the drive (starting at the interface edge). By my quick math, it should be possible to fit 18 cards that way (a 3x3 on each board with the SD cards horizontal and with their contacts towards each other). So 16x4GB is doable, for a 64GB notebook drive.
Figuring $100 for the controller, software development recovery, and connectors and such, plus 16 4GB SD cards at $30 each, I get $580. I will almost guarantee you that prices will fall to this level within 3 months of the first such device being available for much more than this.
I'm not sure what law you're thinking of, but in the United States, that is most definitely not how it works. It's not even remotely close if the Plaintiff is a public figure.
It was mostly intended to be funny. But there were actually several serious points it was trying to make with its humor. The main one is that just because a rule is commonly violated (such as speeding) one cannot just accuse a random person of it. For all I know, you are 11 years old and have never driven a car.
Then I agree with you. The flipside to my argument that they will last about 15 years with normal use is that if you run them at full blast with just write, they will only last about 1.5 years. Now I can't think of any reason anyone would write all that data and never read it back, but there certainly could be applications that this type of memory is unsuited for.
I would like to see the specifications for drives of this type include the cell size, number of erase cycles per cell, the type of wear leveling used, and the forecast write endurance under at least worst case and typical scenarios.
There is no read limit. The write limit is about 100,000 writes (really erasures) per cell.
These devices will have wear leveling. That means that if a cell is close to running out of erase cycles, the drive will move data that has not changed in a very long time into that cell. A few cells will be kept as spares in case some cells don't last as long as they are predicted to.
If you do the math, and figure a typical use scenario as a laptop's primary drive, you get that these drives should outlast mechanical hard drives by many years. For example, a 64GB hard drive with an endurance of 100,000 writes should be able to tolerate about 5 million GB of writes before it fails due to wear.
How long it will take you to run that out depends on your average write rate. But with a reasonable rate (10MB/s) that works out to about 15 years.
Speed doesn't matter. With 16 low-speed cards, you can make one really high-speed one. It just takes a smart controller.
In any event, even if they are slow, the speed limit doesn't come from the flash chips themselves. The speed limit comes from the controller.
This drive has a controller and some flash chips. The cost of the controller is, maybe, $50 tops. The question is -- how much do the flash chips cost? If you can get 4GB flash cards for $24, that means the flash chips inside there must cost at most $24. The means you can sell 64GB of flash chips for $384 without losing your shirt. This even includes the cost of a controller, packaging, and the cost to advertise, stock, and sell the product that we don't need.
There is no rational reason this drive should sell for more than $500, except that there is only limited supply. As soon as supply ramps up, the price will drop to about this value. I'd guess this will take 3 months or so.
First, you mean MBps. We're talking bytes, not bits.
Second, your hard drive can sustain 60MB/s on the fastest part of the drive. Its average is probably much less than that (due to different linear speeds on the inside and outside of the platters).
That speed drops catastrophically in many real-world scenarios. Small random reads, for example, become dominated by seek time and rotational latency and the high transfer rate doesn't help very much. Small random writes are only slightly better.
It is really not "only double". It has a real-world speed that is about twice a high-end hard drive's theoretical maximum speed.
As they come from a chronic speeder who fails to take seriously our great nation's speed limit laws, I see no reason to consider your argument seriously.
"I say: Kryten250 did, at some unspecified time and place, suck a donkey's balls. You then sue me. That makes you the plaintiff. Can you prove that you didn't, ever, suck donkey balls?"
There's probably a name for this logical fallacy.
Men play hockey. Socrates never played hockey. Therefore Socrates is not a man.
That the Plaintiff doesn't have to prove that there is no possible way the statement is true doesn't mean the plaintiff doesn't have the burden of proof.
> If they can't identify "subscribers", how can they pass along DMCA complaints > or terminate the accounts of repeat offenders? If they can't do those things, > does that eliminate their Safe Harbor status?
Short answer, "no".
Slightly longer answer, many services can't do that. Requiring them to would place an impossible burden on anonymous speech.
It is analogous to the question of whether a particular person can solve math problems of a particular type, with some example word problems given. The person does not speak English, so needs a translator to translate the question from English to Spanish. So long as the translator doesn't do any math, the person can solve the problems. If the translator does the math, the person cannot. The question of whether or not the translator does any math may not be simple.
That is precisely what happened in this case. The question was looked into, and it was deemed that the encoder (analogous to the translator) didn't solve the problem but merely translated it into a form the machine itself could solve. Thus, the existence of the encoder proves that the machine can solve the problem.
However, if you disagree about what the encoder does, you will not find the proof convincing.
Nonsense. There are $2,000 video cards today, as there were five years ago. If the "top" means the 95% of sales, it's gone down from 400 2002 dollars (about 500 2007 dollars) to 300 2007 dollars, a drop of 40%.
This is even ignoring the fact that you get so much more for your money. This is just looking at how much less money you're spending.
Looking at how much I've spent and how much I've gotten over the years, buying pretty consistently at about to 90% mark (10% of computers cost more than mine, 90% cost less), I spent $5,000 in 1995 and $2,200 in 2007. That's a 75% drop once you correct for inflation. That's not even comparing the performance and capability.
How would you feel if your local retail store advertised that soda was "$2 for whites, $1.50 for blacks". They might even be able to sell more sodas that way, if you assume the average black person has less disposable income than the average white person. So it might even be a smart business decision. What would you say of a group of black people who resold their sodas for $1.75 outside the store? Would the soda store owner be justified in busting up their operation?
This is precisely what Valve did and is doing. They reduced the price for a group of people on the rationale that they could not afford to pay as much. Some of those people opted to resell for more. Well, duh. That's why price discrimination doesn't work.
The difference is, valve was able to find a way to poison the soda. That doesn't make it right though. Not even if they put a sticker on the soda that said, "for the exclusive consumption of white people".
How does throwing away an upgrade sale and doing nothing whatsoever to harm the vendor that screwed them constitute "protecting their money"? It looks to me more like throwing good money after bad or cutting their nose to spite their face.
Actually, it's much more like a phone call. In the middle of the conversation, you get randomly disconnected. You call back, and continue from approximately where you left off, with some information lost due to the random disconnection. When you finally get back to where you were, you get disconnected again. Eventually, you are able to communicate the information.
Comcast is just being deceptive here. The question is "what are you doing" and they answer "this is the effect we expect what we are doing to have". That doesn't answer the question. Comcast, what are you *DOING*. Not "what effect does what you are doing have on your customers", what are you *DOING* to the *packets*?
"How do you propose that Best Buy creates a mechanism that simultaneously gives managers an incentive to perform, and keeps them from committing costly law violations?"
Four tools do this:
1) Set policy. That is, make it absolutely clear what is and is not permitted. Be clear and consistent.
2) Create incentives not just for most sales but also for fewest complaints and best compliance with the rules.
3) Audit. Call up customers who buy things like extended warranties and ask them if they knew they were buying them. Make sure the rules are followed.
4) Fix problems when they are discovered. If this means changing the policy, change the policy. If this means finding new ways to ensure compliance with existing policy, find them. If you have to fire people, do that.
It's not easy, but it's not *that* hard. In this case, there's a blurry line between just saying "Oh, you get six months free MSN but you have to pay for it after that let me just set that up for you." and "Do you want a free six month trial of MSN? There's no fee if you cancel."
And if you think that even that's not enough disclosure, then *you* tell the managers what must be disclosed. That decision has to be made as store policy, not by each manager or salesman. IMO, nothing but "Would you like a free six month trial of MSN. Your credit card will be billed automatically if you do not cancel within six months" is good enough.
Do we know that the boards weren't coated? In this case, the problem was condensation on connectors, right? It's entirely possible that the boards were coated. I know there are such things as gas-tight connectors, but sometimes they aren't quite as gas tight as they're supposed to be either.
Also, the idea that management rejected the idea of more protection against condensation seems to be pure speculation on your part. Sure, you can blame management for any mistake and you're probably right, but it's not like we know that they specifically rejected changes that would have made for more corrosion resistance. For all we know, they pushed for more and more corrosion resistance until the engineers were sick of it. (Doubtful, of course, but we don't know.)
"It is dismaying that after decades of experience with manned space stations, Russian space engineers still couldn't keep unwanted condensation at bay."
That's a bunch of crap. That's like saying it's dismaying that McDonald's has served billions of burgers and still can't figure out how to make them healthy.
Condensation is "still" a problem because it's one of the big and tricky ones. To get rid of the condensation, you have to get rid of the people.
I've been using Vista on a secondary machine for a few months now. It's a very high-end box, quad-core CPU, 3GB of RAM, 8800 GTS video card, and of course Vista runs great on it. What wouldn't?
I've had a few minor issues with Vista, most of which Microsoft has already offered fixes for. I haven't found anything significantly better, but other than the issues that have been fixed, I haven't found anything worse either.
I'm not sure why people would want XP over Vista, unless they need to support old hardware that they already have or are buying computers under about $700 total price.
Of course, there's a vibrant low-end computer market, and the lack of XP availability would probably hurt it quite a bit. Or push people to use Linux. We'll see.
Perhaps the unavailability of XP will push the $1,000 computer market to Windows? Consider also that the cost of the OS is proportionately higher at the lower end. That extra $130 from not buying Windows could mean a quad-core CPU instead of a dual or 2GB instead of 1GB for your Linux box.
It's not. It's nice that they fixed it, but it wasn't there bug. Firefox, and other programs, were passing invalid URLs from untrusted sources to the operating system.
Secunia most certainly agrees with me that it's not an IE flaw. The page you cited says that the bug affects Firefox. How can an IE vulnerability affect Firefox?
However, looking at the details referenced from that page, it's not quite so clear who is responsible. It's a judgment call. This could be considered either an OS bug or a browser bug depending.
I would argue that it's the browser's job to sanity-check the URL before handing it to the OS. However, if the OS is going to process URLs (and everyone knows URLs sometimes come from untrusted sources), the OS should have a way to denote that an URL is untrusted so that malformed URLs can be rejected.
Firefox's choice to let the OS launch the URL is, IMO, disastrous. Firefox should launch the URL itself if the URL comes from an untrusted source.
Neither. Parameter validation is a common source of bugs in many APIs where one program launches another. Regardless of the API specification, every program must sanity check all of its invocation parameters.
Any program that is intended to be launched from a browser is going to be launched with untrusted parameters. This means that they have to validate them. There's just no way for the browser to know what parameters are valid for Adobe Reader or Macromedia Flash.
These are programs that were designed to be launched with untrusted parameters. They have to validate every single one of them completely.
From what I understand, and there isn't much in the way of technical details available, this is not an IE flaw. IE, correctly, doesn't assume that a URI is invalid just because it looks odd. This is correct, because there is no way IE can know if an URI for another protocol is valid or invalid. It is the responsibility of the target program to sanitize its input, knowing full well that it comes from an untrusted source.
"Apple is also perfectly within their rights to not give warranty service to those that modded their phone. The Magnuson-Moss Act only provides protection to those whose aftermarket bits did not cause the phone to die. If these folks had not modded their phone, the update would not have killed it."
Umm, the modification didn't cause the phone to die, the update did. The phone worked after the mod, it didn't work after the update.
Isn't it obvious that Apple released the update at least knowing that it would brick a modded phone? Isn't it even probable that they intended this?
Apple's choice bricked the phone. Apple's actions were intended to brick modded phones.
Denying warranty coverage on this basis is *EXACTLY* what Magnusson-Moss prohibits.
Slashdot Mirror
A 1U box with 128 such slots should be quite doable. A 3.5 inch hard drive sized box with 24 such slots should be quite doable. SATA2 would probably be the ideal interface.
There are really only three pieces that you need. First you need the main controller that talks SATA2 out one end and talks to the flash controllers on the other end. Then you need the interface hardware to connect to so many SD slots. Then you need the software to do the wear leveling and drive emulation.
Notebook size is a bit trickier. A full-sized SD card is 24mm by 32mm and is 2.1mm thick. A typical notebook drive (2.5 inch) is 100mm by 70mm and 9.5mm thick. You will need some space for the housing and interface electronics and you can't stand the SD cards up (assuming you use full-sized SD cards). One possibility would be to have SD slots on both sides of two long narrow vertical boards that run down the center of the drive (starting at the interface edge). By my quick math, it should be possible to fit 18 cards that way (a 3x3 on each board with the SD cards horizontal and with their contacts towards each other). So 16x4GB is doable, for a 64GB notebook drive.
Figuring $100 for the controller, software development recovery, and connectors and such, plus 16 4GB SD cards at $30 each, I get $580. I will almost guarantee you that prices will fall to this level within 3 months of the first such device being available for much more than this.
I'm not sure what law you're thinking of, but in the United States, that is most definitely not how it works. It's not even remotely close if the Plaintiff is a public figure.
It was mostly intended to be funny. But there were actually several serious points it was trying to make with its humor. The main one is that just because a rule is commonly violated (such as speeding) one cannot just accuse a random person of it. For all I know, you are 11 years old and have never driven a car.
Then I agree with you. The flipside to my argument that they will last about 15 years with normal use is that if you run them at full blast with just write, they will only last about 1.5 years. Now I can't think of any reason anyone would write all that data and never read it back, but there certainly could be applications that this type of memory is unsuited for.
I would like to see the specifications for drives of this type include the cell size, number of erase cycles per cell, the type of wear leveling used, and the forecast write endurance under at least worst case and typical scenarios.
There is no read limit. The write limit is about 100,000 writes (really erasures) per cell.
These devices will have wear leveling. That means that if a cell is close to running out of erase cycles, the drive will move data that has not changed in a very long time into that cell. A few cells will be kept as spares in case some cells don't last as long as they are predicted to.
If you do the math, and figure a typical use scenario as a laptop's primary drive, you get that these drives should outlast mechanical hard drives by many years. For example, a 64GB hard drive with an endurance of 100,000 writes should be able to tolerate about 5 million GB of writes before it fails due to wear.
How long it will take you to run that out depends on your average write rate. But with a reasonable rate (10MB/s) that works out to about 15 years.
Speed doesn't matter. With 16 low-speed cards, you can make one really high-speed one. It just takes a smart controller.
In any event, even if they are slow, the speed limit doesn't come from the flash chips themselves. The speed limit comes from the controller.
This drive has a controller and some flash chips. The cost of the controller is, maybe, $50 tops. The question is -- how much do the flash chips cost? If you can get 4GB flash cards for $24, that means the flash chips inside there must cost at most $24. The means you can sell 64GB of flash chips for $384 without losing your shirt. This even includes the cost of a controller, packaging, and the cost to advertise, stock, and sell the product that we don't need.
There is no rational reason this drive should sell for more than $500, except that there is only limited supply. As soon as supply ramps up, the price will drop to about this value. I'd guess this will take 3 months or so.
First, you mean MBps. We're talking bytes, not bits.
Second, your hard drive can sustain 60MB/s on the fastest part of the drive. Its average is probably much less than that (due to different linear speeds on the inside and outside of the platters).
That speed drops catastrophically in many real-world scenarios. Small random reads, for example, become dominated by seek time and rotational latency and the high transfer rate doesn't help very much. Small random writes are only slightly better.
It is really not "only double". It has a real-world speed that is about twice a high-end hard drive's theoretical maximum speed.
As they come from a chronic speeder who fails to take seriously our great nation's speed limit laws, I see no reason to consider your argument seriously.
"I say: Kryten250 did, at some unspecified time and place, suck a donkey's balls.
You then sue me. That makes you the plaintiff. Can you prove that you didn't, ever, suck donkey balls?"
There's probably a name for this logical fallacy.
Men play hockey.
Socrates never played hockey.
Therefore Socrates is not a man.
That the Plaintiff doesn't have to prove that there is no possible way the statement is true doesn't mean the plaintiff doesn't have the burden of proof.
> If they can't identify "subscribers", how can they pass along DMCA complaints
> or terminate the accounts of repeat offenders? If they can't do those things,
> does that eliminate their Safe Harbor status?
Short answer, "no".
Slightly longer answer, many services can't do that. Requiring them to would place an impossible burden on anonymous speech.
It is analogous to the question of whether a particular person can solve math problems of a particular type, with some example word problems given. The person does not speak English, so needs a translator to translate the question from English to Spanish. So long as the translator doesn't do any math, the person can solve the problems. If the translator does the math, the person cannot. The question of whether or not the translator does any math may not be simple.
That is precisely what happened in this case. The question was looked into, and it was deemed that the encoder (analogous to the translator) didn't solve the problem but merely translated it into a form the machine itself could solve. Thus, the existence of the encoder proves that the machine can solve the problem.
However, if you disagree about what the encoder does, you will not find the proof convincing.
There is so much wrong with your post, I don't even know where to begin.
> CPUs: 5 years ago, ~$1k was top, ~$300 mid-line, ~$125 low-end. Today, same.
Really? low-end CPUs are 125 2002 dollars today? I don't think so.
> Video: 5 years ago, ~$400 was top, ~$150 mid-line, ~$50 low-end. Today, it's gone up. ~$700 top, ~$300
> mid-line, ~$100 low-end.
Nonsense. There are $2,000 video cards today, as there were five years ago. If the "top" means the 95% of sales, it's gone down from 400 2002 dollars (about 500 2007 dollars) to 300 2007 dollars, a drop of 40%.
This is even ignoring the fact that you get so much more for your money. This is just looking at how much less money you're spending.
Looking at how much I've spent and how much I've gotten over the years, buying pretty consistently at about to 90% mark (10% of computers cost more than mine, 90% cost less), I spent $5,000 in 1995 and $2,200 in 2007. That's a 75% drop once you correct for inflation. That's not even comparing the performance and capability.
DS
How would you feel if your local retail store advertised that soda was "$2 for whites, $1.50 for blacks". They might even be able to sell more sodas that way, if you assume the average black person has less disposable income than the average white person. So it might even be a smart business decision. What would you say of a group of black people who resold their sodas for $1.75 outside the store? Would the soda store owner be justified in busting up their operation?
This is precisely what Valve did and is doing. They reduced the price for a group of people on the rationale that they could not afford to pay as much. Some of those people opted to resell for more. Well, duh. That's why price discrimination doesn't work.
The difference is, valve was able to find a way to poison the soda. That doesn't make it right though. Not even if they put a sticker on the soda that said, "for the exclusive consumption of white people".
How does throwing away an upgrade sale and doing nothing whatsoever to harm the vendor that screwed them constitute "protecting their money"? It looks to me more like throwing good money after bad or cutting their nose to spite their face.
Actually, it's much more like a phone call. In the middle of the conversation, you get randomly disconnected. You call back, and continue from approximately where you left off, with some information lost due to the random disconnection. When you finally get back to where you were, you get disconnected again. Eventually, you are able to communicate the information.
Comcast is just being deceptive here. The question is "what are you doing" and they answer "this is the effect we expect what we are doing to have". That doesn't answer the question. Comcast, what are you *DOING*. Not "what effect does what you are doing have on your customers", what are you *DOING* to the *packets*?
"How do you propose that Best Buy creates a mechanism that simultaneously gives managers an incentive to perform, and keeps them from committing costly law violations?"
Four tools do this:
1) Set policy. That is, make it absolutely clear what is and is not permitted. Be clear and consistent.
2) Create incentives not just for most sales but also for fewest complaints and best compliance with the rules.
3) Audit. Call up customers who buy things like extended warranties and ask them if they knew they were buying them. Make sure the rules are followed.
4) Fix problems when they are discovered. If this means changing the policy, change the policy. If this means finding new ways to ensure compliance with existing policy, find them. If you have to fire people, do that.
It's not easy, but it's not *that* hard. In this case, there's a blurry line between just saying "Oh, you get six months free MSN but you have to pay for it after that let me just set that up for you." and "Do you want a free six month trial of MSN? There's no fee if you cancel."
And if you think that even that's not enough disclosure, then *you* tell the managers what must be disclosed. That decision has to be made as store policy, not by each manager or salesman. IMO, nothing but "Would you like a free six month trial of MSN. Your credit card will be billed automatically if you do not cancel within six months" is good enough.
Do we know that the boards weren't coated? In this case, the problem was condensation on connectors, right? It's entirely possible that the boards were coated. I know there are such things as gas-tight connectors, but sometimes they aren't quite as gas tight as they're supposed to be either.
Also, the idea that management rejected the idea of more protection against condensation seems to be pure speculation on your part. Sure, you can blame management for any mistake and you're probably right, but it's not like we know that they specifically rejected changes that would have made for more corrosion resistance. For all we know, they pushed for more and more corrosion resistance until the engineers were sick of it. (Doubtful, of course, but we don't know.)
Absolutely.
"It is dismaying that after decades of experience with manned space stations, Russian space engineers still couldn't keep unwanted condensation at bay."
That's a bunch of crap. That's like saying it's dismaying that McDonald's has served billions of burgers and still can't figure out how to make them healthy.
Condensation is "still" a problem because it's one of the big and tricky ones. To get rid of the condensation, you have to get rid of the people.
Do you have any evidence? Or do you think that everyone who disagrees with you must be a shill?
I've been using Vista on a secondary machine for a few months now. It's a very high-end box, quad-core CPU, 3GB of RAM, 8800 GTS video card, and of course Vista runs great on it. What wouldn't?
I've had a few minor issues with Vista, most of which Microsoft has already offered fixes for. I haven't found anything significantly better, but other than the issues that have been fixed, I haven't found anything worse either.
I'm not sure why people would want XP over Vista, unless they need to support old hardware that they already have or are buying computers under about $700 total price.
Of course, there's a vibrant low-end computer market, and the lack of XP availability would probably hurt it quite a bit. Or push people to use Linux. We'll see.
Perhaps the unavailability of XP will push the $1,000 computer market to Windows? Consider also that the cost of the OS is proportionately higher at the lower end. That extra $130 from not buying Windows could mean a quad-core CPU instead of a dual or 2GB instead of 1GB for your Linux box.
It could happen.
It's not. It's nice that they fixed it, but it wasn't there bug. Firefox, and other programs, were passing invalid URLs from untrusted sources to the operating system.
Secunia most certainly agrees with me that it's not an IE flaw. The page you cited says that the bug affects Firefox. How can an IE vulnerability affect Firefox?
However, looking at the details referenced from that page, it's not quite so clear who is responsible. It's a judgment call. This could be considered either an OS bug or a browser bug depending.
I would argue that it's the browser's job to sanity-check the URL before handing it to the OS. However, if the OS is going to process URLs (and everyone knows URLs sometimes come from untrusted sources), the OS should have a way to denote that an URL is untrusted so that malformed URLs can be rejected.
Firefox's choice to let the OS launch the URL is, IMO, disastrous. Firefox should launch the URL itself if the URL comes from an untrusted source.
Neither. Parameter validation is a common source of bugs in many APIs where one program launches another. Regardless of the API specification, every program must sanity check all of its invocation parameters.
Any program that is intended to be launched from a browser is going to be launched with untrusted parameters. This means that they have to validate them. There's just no way for the browser to know what parameters are valid for Adobe Reader or Macromedia Flash.
These are programs that were designed to be launched with untrusted parameters. They have to validate every single one of them completely.
From what I understand, and there isn't much in the way of technical details available, this is not an IE flaw. IE, correctly, doesn't assume that a URI is invalid just because it looks odd. This is correct, because there is no way IE can know if an URI for another protocol is valid or invalid. It is the responsibility of the target program to sanitize its input, knowing full well that it comes from an untrusted source.
"Apple is also perfectly within their rights to not give warranty service to those that modded their phone. The Magnuson-Moss Act only provides protection to those whose aftermarket bits did not cause the phone to die. If these folks had not modded their phone, the update would not have killed it."
Umm, the modification didn't cause the phone to die, the update did. The phone worked after the mod, it didn't work after the update.
Isn't it obvious that Apple released the update at least knowing that it would brick a modded phone? Isn't it even probable that they intended this?
Apple's choice bricked the phone. Apple's actions were intended to brick modded phones.
Denying warranty coverage on this basis is *EXACTLY* what Magnusson-Moss prohibits.