Yes. The _only_ way to have a usable backup of that data is to have a second RAID array.
If you just want a backup in order to check a box on a list (like most companies), then tape might work.
"normalization of deviance" is what caused the problems. I can see fundamentalists having a field day with that one.
Actually looking directly at the problem is the only way to fix it ultimately.
I like Bob Lewis' take on investigations in a blog he wrote about NASA vs other government Agencies. http://www.issurvivor.com/shop...
Actually, I suspect most of the folks who end up buying the phone will be buying them specifically to figure out how to break them.
It's a brilliant marketing scheme. Justin Bieber and Kim Jong Un will each buy one for security reasons and the other 70,000 Boeing ends up selling will all go to security researchers in China, Russia and Europe.
neither resistance nor prevention is the goal. The goal is to prevent un-noticeable tampering.
If you get your phone back from the lost and found at the local Chinese restaurant, you want to make sure they didn't copy the sim card so-to-speak.
This phone is designed for the sorts of people who build and defend against things like Stuxnet.
I'm too am not convinced a 4-10C will be death of humanity.
However, losing 90% of the species means we lose a huge part of the ecosystem, and we depend on that ecosystem for far more than most folks understand.
Part two is that the world is already full of people. If anyone wants to migrate, they'll have to fight for it--which is not unknown to happen.
imo, most all the huge wars in the past have happened because of too many people and too few resources. Look forward to more as resources die off.
Or maybe Glass will be another Segway, theoretically useful and cost-effective but socially unfashionable and dorky.
Fashion counts for widespread social acceptance even if the average slashdaughter pretends to not care about fashion.
Mathematically, the odds of winning are so small they equal zero.
I explain this by saying I've won $20 twice in the lottery without ever entering.
Once I received a lottery ticket from a Realtor (a somewhat standard marketing technique) and I've gotten them as prizes at work contests. I've never bought a lottery ticket because my chances of winning don't actually change.
imo,l the difference between statistics and probability is that every customer in a casino is gambling, but the casino itself is most definitely not gambling. It is the difference between owning one hand of cards and owning ten thousand hands.
Roll a dice. Each of the outcomes only has a probability of 16.p6 % (assuming a fair d6), which is fairly unlikely. Yet, there's a 100 % probability that you will obtain one of these unlikely results.
unless one die rolls into a crack or becomes a leaner against the table or rolls under the refrigerator in which case there is a result but it's more along the lines of Schrodinger's cat or Christ could come back to earth and take up your friends in rapture along with the dice and leave you hanging or the universe could explode. I'd give it five 9s instead. 99.999%
As far as the old article is concerned, the problem was that the Lottery commission, in order to maintain sales, interfered with the actual randomization. Every pack of 1,000 tickets sent to a store has so many $2, $5, and $20 winners. A clerk at the store paying attention would open a new pack of a thousand tickets and keep track of the winners. If there were fewer than expected, then it actually made sense to buy the last 150 tickets of the pack (using friends and accomplices) This is best done at the sorts of run-down liquor stores where no one takes a lottery ticket anyplace else..
It wasn't serial numbers. At first glance, it is poor management decision to alter the odds. However, that increases sales which is why slot machines actually pay off something regularly.And if your job is to increase sales of packs of lottery tickets to stores, lots of winners spread thru _every_ store make a lot of sense.
I think one of the issues we are missing when teaching stats is that the bell curve (Gaussian distbtn) and the long tail (1/t curve) are both measures of populations.
Consider gun ownership in the US
There is approximately one gun per person in the nation; however _most_ people, more than half, do not own a gun. This situation is modeled like most any other unequal one (but it also matches the atomic configuration of atoms during a phase change). Most folks don't own a gun. The next largest group is those who have one gun. Fewer folks own 2, then 3, then 4. And there are a few folks who own tens or hundreds of guns (and we don't count 101, 102, 103 when performing the measurements).
The bell curve and the long tail curve are just different sides of the same coin (I hope that analogy doesn't confuse anyone;-)
One of the issues is the conflation of time with probability. A coin flip is half odds whether you flip it every second or every ten thousand years.
However, when you flip a coin every second for ten thousand years,you get different results. The million monkeys typing a million years to produce Shakespeare is a perfect example of how multiplying probability by time is like dividing integers by zero. Things get funky.
If the Fukushima risk analysis looks at one event per day versus one event per hour or one event per millisecond, you get different results for the 'same' amount of risk.
imo, it's why the Drake equations calculating the probability of life in the universe to be almost certain are almost certainly wrong. They calculated star formation PER YEAR for the entire age of the universe. That gives you a lot of events to sift through. A more accurate approach would be to avoid introducing time into the calculations and instead count the number of stars in the universe (10 to the 22nd power) and then realize that translates into life being unique in the entire universe if there are 22 events with a 1 out of ten chance of occurring.
Must be because you Like it and want to romance it.
But from the article, it sounds like MarcZ already knows how much you want to make love to his software.
As long as it's a legal contract it can be as 'unfair' as you can get them to sign.
uh.... no. an unfair or unenforceable or one-sided EULA is not a legal contract even if I click it.
The entire idea of enforcing contracts is to make _society_ run better, not to choose winners that make most everybody else losers.
Hence fairness is a legal concept often enforced.
Hah, its funny you say that, because I ran into a similar situation working for a somewhat niche but well-known and respected website with many established customers. What happened was that we were bought out by an obscure advertising shit-peddler, and immediately came the MBAs telling us about web 2.0 and how the things on our site were rendered too small. One of the sonofabitches actually said, " When people see small things on your screen, they think small. Think BIG! 16-point text and 500-pixel padding minimum! "
It was clear that their idea of a redesign was here to stay, but we stalled its implementation as much as we could -- well, except for the Jewish editors whose eyes always had dollar-signs dancing in 'em -- and when we finally rolled it out the users hated it so much that constructive criticism wasn't enough - established users with excellent karma(a term for measurement of a member's positive participation) not only badmouthed the redesign, but used their moderation points moderating up the trolls(people who disparage, often using profanity) who did likewise. It go to be so bad that they all organized a boycott, which is going on this week.
Man, if I weren't boycotting slashdot this week, I would mod you up.
Yes. The _only_ way to have a usable backup of that data is to have a second RAID array.
If you just want a backup in order to check a box on a list (like most companies), then tape might work.
It's so sad and pathetic that the metric being used by people is amount of Facebook "friends".
You mean as opposed to believing people are prettier and more interesting in real life because of all their 'real' "friends."?
I "like" you Ronald McDonald.
Yes, there are certainly some companies who pay to find out why people flock to a specific brand.
And maybe measurements are the entire problem with the filter.
You know, mixing quarts and litres like they mixed metres and yards on some mission.
Well if you were from Burma, you wouldn't have been confused.
"normalization of deviance" is what caused the problems. I can see fundamentalists having a field day with that one.
Actually looking directly at the problem is the only way to fix it ultimately.
I like Bob Lewis' take on investigations in a blog he wrote about NASA vs other government Agencies.
http://www.issurvivor.com/shop...
Actually, I suspect most of the folks who end up buying the phone will be buying them specifically to figure out how to break them.
It's a brilliant marketing scheme. Justin Bieber and Kim Jong Un will each buy one for security reasons and the other 70,000 Boeing ends up selling will all go to security researchers in China, Russia and Europe.
neither resistance nor prevention is the goal. The goal is to prevent un-noticeable tampering.
If you get your phone back from the lost and found at the local Chinese restaurant, you want to make sure they didn't copy the sim card so-to-speak.
This phone is designed for the sorts of people who build and defend against things like Stuxnet.
I really doubt it is actually meant to blow itself up though.
If they used the right kind of battery it could ;-)
Why is it so very hard for people to accept that increases in CO2 in the atmosphere, whatever their source, is not a good thing for a lot of species?
oft times it is because they believe their livelihood depends on their not understanding.
I'm too am not convinced a 4-10C will be death of humanity.
However, losing 90% of the species means we lose a huge part of the ecosystem, and we depend on that ecosystem for far more than most folks understand.
Part two is that the world is already full of people. If anyone wants to migrate, they'll have to fight for it--which is not unknown to happen.
imo, most all the huge wars in the past have happened because of too many people and too few resources. Look forward to more as resources die off.
Or maybe Glass will be another Segway, theoretically useful and cost-effective but socially unfashionable and dorky.
Fashion counts for widespread social acceptance even if the average slashdaughter pretends to not care about fashion.
Sounds like there's a lot of Glass owners who don't know what "socially inappropriate" means. Hence the etiquette list?
Even more absurd is that we're talking about proper social etiquette on slashdot. ha ha ha.
What sort of tie should I wear to a Justin Bieber concert?
Then why are there no unicorns?
Because Noah couldn't get them on board. They were out in the rain playing silly games.
The Unicorn by Tom Lehrer
- The worst aspect of intelligence is being trapped by it
Being trapped by intelligence is the best aspect. ;-)
I think
Mathematically, the odds of winning are so small they equal zero.
I explain this by saying I've won $20 twice in the lottery without ever entering.
Once I received a lottery ticket from a Realtor (a somewhat standard marketing technique) and I've gotten them as prizes at work contests. I've never bought a lottery ticket because my chances of winning don't actually change.
imo,l the difference between statistics and probability is that every customer in a casino is gambling, but the casino itself is most definitely not gambling. It is the difference between owning one hand of cards and owning ten thousand hands.
Roll a dice. Each of the outcomes only has a probability of 16.p6 % (assuming a fair d6), which is fairly unlikely. Yet, there's a 100 % probability that you will obtain one of these unlikely results.
unless one die rolls into a crack or becomes a leaner against the table or rolls under the refrigerator in which case there is a result but it's more along the lines of Schrodinger's cat or Christ could come back to earth and take up your friends in rapture along with the dice and leave you hanging or the universe could explode. I'd give it five 9s instead. 99.999%
As far as the old article is concerned, the problem was that the Lottery commission, in order to maintain sales, interfered with the actual randomization. Every pack of 1,000 tickets sent to a store has so many $2, $5, and $20 winners. A clerk at the store paying attention would open a new pack of a thousand tickets and keep track of the winners. If there were fewer than expected, then it actually made sense to buy the last 150 tickets of the pack (using friends and accomplices) This is best done at the sorts of run-down liquor stores where no one takes a lottery ticket anyplace else. .
It wasn't serial numbers. At first glance, it is poor management decision to alter the odds. However, that increases sales which is why slot machines actually pay off something regularly.And if your job is to increase sales of packs of lottery tickets to stores, lots of winners spread thru _every_ store make a lot of sense.
I think one of the issues we are missing when teaching stats is that the bell curve (Gaussian distbtn) and the long tail (1/t curve) are both measures of populations.
;-)
Consider gun ownership in the US
There is approximately one gun per person in the nation; however _most_ people, more than half, do not own a gun. This situation is modeled like most any other unequal one (but it also matches the atomic configuration of atoms during a phase change). Most folks don't own a gun. The next largest group is those who have one gun. Fewer folks own 2, then 3, then 4. And there are a few folks who own tens or hundreds of guns (and we don't count 101, 102, 103 when performing the measurements).
The bell curve and the long tail curve are just different sides of the same coin (I hope that analogy doesn't confuse anyone
One of the issues is the conflation of time with probability. A coin flip is half odds whether you flip it every second or every ten thousand years.
However, when you flip a coin every second for ten thousand years,you get different results. The million monkeys typing a million years to produce Shakespeare is a perfect example of how multiplying probability by time is like dividing integers by zero. Things get funky.
If the Fukushima risk analysis looks at one event per day versus one event per hour or one event per millisecond, you get different results for the 'same' amount of risk.
imo, it's why the Drake equations calculating the probability of life in the universe to be almost certain are almost certainly wrong. They calculated star formation PER YEAR for the entire age of the universe. That gives you a lot of events to sift through. A more accurate approach would be to avoid introducing time into the calculations and instead count the number of stars in the universe (10 to the 22nd power) and then realize that translates into life being unique in the entire universe if there are 22 events with a 1 out of ten chance of occurring.
Summary of article: Consider improbable event X. Repeat event X numerous times (10s, hundreds, thousands etc). Suddenly, event X is quite probable.
You mean like a million monkeys typing for a million years will produce Shakespeare?
I just made that example up.
Fuck Facebook.
Must be because you Like it and want to romance it.
But from the article, it sounds like MarcZ already knows how much you want to make love to his software.
As long as it's a legal contract it can be as 'unfair' as you can get them to sign.
uh.... no. an unfair or unenforceable or one-sided EULA is not a legal contract even if I click it.
The entire idea of enforcing contracts is to make _society_ run better, not to choose winners that make most everybody else losers.
Hence fairness is a legal concept often enforced.
Hah, its funny you say that, because I ran into a similar situation working for a somewhat niche but well-known and respected website with many established customers. What happened was that we were bought out by an obscure advertising shit-peddler, and immediately came the MBAs telling us about web 2.0 and how the things on our site were rendered too small. One of the sonofabitches actually said, " When people see small things on your screen, they think small. Think BIG! 16-point text and 500-pixel padding minimum! "
It was clear that their idea of a redesign was here to stay, but we stalled its implementation as much as we could -- well, except for the Jewish editors whose eyes always had dollar-signs dancing in 'em -- and when we finally rolled it out the users hated it so much that constructive criticism wasn't enough - established users with excellent karma(a term for measurement of a member's positive participation) not only badmouthed the redesign, but used their moderation points moderating up the trolls(people who disparage, often using profanity) who did likewise. It go to be so bad that they all organized a boycott, which is going on this week.
Man, if I weren't boycotting slashdot this week, I would mod you up.