What you're really thinking of is Received Pronunciation. Although if you don't want to sound like a foreigner, I'd suggest learning Standard American English instead of one of the mid-Atlantic accents.
At the grad school I went to, if you were foreign, you had to have a language test. If you didn't do well, you had to take classes to help you learn how to speak English more clearly. We had some foreign TAs who still were nearly unintelligible, but that was fairly uncommon and somewhat of an embarrassment to the grad student selection process. Potential professor hires have to give a lecture and have private talks with groups of students, and being unintelligible is generally a deal-killer.
Wood already works for "carbon fixation" and you can make things with it that people will actually keep. My mother has some "fixated carbon" in the living room over 100 years old. Just grow a tree and make a desk.
Apparently the IPCC agrees with you, even. However, relying on wood as a sole means of carbon sequestration requires planting far more trees than we can reasonably dedicate land to.
Planting trees to counteract CO2 emissions is cheap and effective, but it's not enough. We already know how to do it, so you're probably not going to see any news about new advances in tree-planting technology on Slashdot.
Dropbox - for a while they claimed that only user has encryption keys and it is impossible for their staff to decrypt anything
Actually, they never claimed that only the user has the encryption keys, and they didn't say "impossible", they said "can't", which is a softer guarantee.
Just goes to show you when reading security guarantees, "fraud" is not nearly as much a problem as companies being intentionally vague and optimistic, customers reading in to their statements what they want to hear, and everyone having poor reading comprehension and inadequate skepticism.
do EA and Sony have the right to oblige you not to sue them once you click the button
Even though the ToS might say that, that doesn't make it legally binding, and judges often ignore such phrases. I still don't think they ought to be able to give you the impression that it's true, but eh.
All I know is I tried a couple of tools like Recuva on one I ran Eraser on and it got a big fat nothing. I just wasn't sure how much better the tools the feds use are for such things.
Yeah, no commercial-grade, simple tool will work at all even with a single pass of zeroes. SSDs often require two passes because of overprovisioning. But even that can only be detected by removing the flash chips from the SSD and using very unpleasant electronic techniques. (Legally, that means that your methods will be immediately called into question, and reconstructing any useful information will be very time-consuming.)
The Feds mostly have access to more time and well-trained personnel, but honestly, trying even to recover data from bad blocks or host-protected areas is notoriously difficult and unrewarding. It has to be something spectacular for law enforcement to bother trying that.
I agree, the paranoid N-wipe methods just take too long, when any forensics guy will tell you that they're just not needed.
Oh, man! I've totally got to look in to the hybrids. The big problem with weird drive areas (bad sectors, host protected area, overprovisioning on SSDs, and the cache on a hybrid) is that a lot of drives either don't implement or improperly implement the secure-erase ATA commands. But they return "oh yes, I totally did secure-erase". So you're basically guessing unless you do careful (often very difficult) analysis.
Yeah, if you're getting conventional disks, a single pass of zeros is SOP. In the incredibly unlikely event that it becomes "a big deal", the history of the drive is really relevant. Imagine some "evil data" is in one of these rare places that doesn't get zeroed, and then the person who gets the drive is investigated carefully. The testimony that you got the drive, zeroed it once, and then gave it to the accused will cast such serious doubt on any information extracted from those sectors (there's, of course, a perfectly reasonable reason that data would be there) that evidence about what they contain is going to be worthless without a fortune of corroborating evidence. It's quite safe.
Hopefully it's c and not the local speed of light. The fact that neutrinos go faster than the local speed of light is not only well-known, it's one of the standard methods of detecting neutrinos. (Build a water tank in an abandoned mine and watch for Cherenkov radiation.)
Nobody in the field of computer forensics has ever claimed or had a reason to believe that any data is recoverable from modern hard drives that have been wiped with a single pass of zeroes (or any other pattern). The police and FBI certainly don't have the technology -- unless they've never used information gathered that way in court and managed not to tell anyone about it.
There are a few exceptions. First, flash / SSD drives are weird. There's a good paper on it, but the short story is that almost all flash drives can be erased by two passes of zeroes if you're wiping the whole drive. (If you're trying to wipe a single file, you're likely to fail.) Second, a logical pass of zeroing (like what you'd get with dban or dd) does not erase data contained in bad sectors that have been mapped out by the drive's firmware or in the drive's other protected areas. Bad sectors can contain user data and can sometimes be read.
Same as the distinction between malware and software: the user wants one, but not the other.
There's no objective definition of evil software -- there is only whether it is wanted or not by the user. One very reasonable thing the user might want is to cede making that decision to someone who is better able to make that judgement. One very reasonable thing the user might want is to make that judgement for themselves.
Politifact has a good summary of how Social Security is not a Ponzi scheme that applies here: First, you know how the bank works (with respect to reserves, that is). The poker site was defrauding users by stated that they behaved one way and behaving in a different (more profitable) way. Second, banks are generally considered a sustainable business model, whereas the poker site's business model is distinctly unsustainable -- at some point, they will run out of money to pay off winners, even if there is no "run" on the site.
Both man and info can handle that reasonably well -- man does not require a one-to-one association between commands and man pages, and much documentation in man is better than "command line parameters". Organization does get a bit ugly in man, though.
Wow, that is the most obviously-bad curve fitting I have ever seen.
Well, second most. The most obviously bad graphed observed temperatures from 1998 to 2008 and showed a "trend" by drawing a line between the 1998 measurement and the 2008 measurement.
The ballpark observed and predicted warming rate is actually about 1 degree Celcius per century.
Experts don't claim four-figure accuracy on temperature prediction. They don't even claim accurate temperature prediction over short time scales, which is not surprising, given the difficulty of modeling short-term behaviors of chaotic systems.
Fortunately, science and the world at large probably don't give a shit about whether you will listen or your arbitrary requirements.
Slashdot Mirror
Only where half = 27%. Also, OnStar predates the government owning a large share of GM.
What you're really thinking of is Received Pronunciation. Although if you don't want to sound like a foreigner, I'd suggest learning Standard American English instead of one of the mid-Atlantic accents.
At the grad school I went to, if you were foreign, you had to have a language test. If you didn't do well, you had to take classes to help you learn how to speak English more clearly. We had some foreign TAs who still were nearly unintelligible, but that was fairly uncommon and somewhat of an embarrassment to the grad student selection process. Potential professor hires have to give a lecture and have private talks with groups of students, and being unintelligible is generally a deal-killer.
We were dumbfounded.
It was actually an experiment to see how long it would take for someone to say, "We can't read your handwriting."
From my experience as a TA, I'd guess that none of you ever said anything, except maybe on your course evaluations at the end. Good job!
Wood already works for "carbon fixation" and you can make things with it that people will actually keep. My mother has some "fixated carbon" in the living room over 100 years old. Just grow a tree and make a desk.
Apparently the IPCC agrees with you, even. However, relying on wood as a sole means of carbon sequestration requires planting far more trees than we can reasonably dedicate land to.
Planting trees to counteract CO2 emissions is cheap and effective, but it's not enough. We already know how to do it, so you're probably not going to see any news about new advances in tree-planting technology on Slashdot.
Dropbox - for a while they claimed that only user has encryption keys and it is impossible for their staff to decrypt anything
Actually, they never claimed that only the user has the encryption keys, and they didn't say "impossible", they said "can't", which is a softer guarantee.
Just goes to show you when reading security guarantees, "fraud" is not nearly as much a problem as companies being intentionally vague and optimistic, customers reading in to their statements what they want to hear, and everyone having poor reading comprehension and inadequate skepticism.
do EA and Sony have the right to oblige you not to sue them once you click the button
Even though the ToS might say that, that doesn't make it legally binding, and judges often ignore such phrases. I still don't think they ought to be able to give you the impression that it's true, but eh.
All I know is I tried a couple of tools like Recuva on one I ran Eraser on and it got a big fat nothing. I just wasn't sure how much better the tools the feds use are for such things.
Yeah, no commercial-grade, simple tool will work at all even with a single pass of zeroes. SSDs often require two passes because of overprovisioning. But even that can only be detected by removing the flash chips from the SSD and using very unpleasant electronic techniques. (Legally, that means that your methods will be immediately called into question, and reconstructing any useful information will be very time-consuming.)
The Feds mostly have access to more time and well-trained personnel, but honestly, trying even to recover data from bad blocks or host-protected areas is notoriously difficult and unrewarding. It has to be something spectacular for law enforcement to bother trying that.
I agree, the paranoid N-wipe methods just take too long, when any forensics guy will tell you that they're just not needed.
Oh, man! I've totally got to look in to the hybrids. The big problem with weird drive areas (bad sectors, host protected area, overprovisioning on SSDs, and the cache on a hybrid) is that a lot of drives either don't implement or improperly implement the secure-erase ATA commands. But they return "oh yes, I totally did secure-erase". So you're basically guessing unless you do careful (often very difficult) analysis.
Yeah, if you're getting conventional disks, a single pass of zeros is SOP. In the incredibly unlikely event that it becomes "a big deal", the history of the drive is really relevant. Imagine some "evil data" is in one of these rare places that doesn't get zeroed, and then the person who gets the drive is investigated carefully. The testimony that you got the drive, zeroed it once, and then gave it to the accused will cast such serious doubt on any information extracted from those sectors (there's, of course, a perfectly reasonable reason that data would be there) that evidence about what they contain is going to be worthless without a fortune of corroborating evidence. It's quite safe.
Proof? I don't think you're an experimental physicist at all!
You're mostly right, but I have to say, I'm so happy that someone is able to properly interpret "as far as we can tell" on Slashdot.
Hopefully it's c and not the local speed of light. The fact that neutrinos go faster than the local speed of light is not only well-known, it's one of the standard methods of detecting neutrinos. (Build a water tank in an abandoned mine and watch for Cherenkov radiation.)
Depends on the frequency. Low-frequency radio waves propagate reasonably well through the ground.
As far as we can tell, c.
Nobody in the field of computer forensics has ever claimed or had a reason to believe that any data is recoverable from modern hard drives that have been wiped with a single pass of zeroes (or any other pattern). The police and FBI certainly don't have the technology -- unless they've never used information gathered that way in court and managed not to tell anyone about it.
There are a few exceptions. First, flash / SSD drives are weird. There's a good paper on it, but the short story is that almost all flash drives can be erased by two passes of zeroes if you're wiping the whole drive. (If you're trying to wipe a single file, you're likely to fail.) Second, a logical pass of zeroing (like what you'd get with dban or dd) does not erase data contained in bad sectors that have been mapped out by the drive's firmware or in the drive's other protected areas. Bad sectors can contain user data and can sometimes be read.
Naturally, Mr. Casey may have a different view!/quote.
He does not. He also pronounces it "Owen".
That'd work if the planet was a lot smaller.
Just like government-run water, sewer, and electricity, right?
In this town, our municipal trash collection is pay-per-use as well, actually.
Same as the distinction between malware and software: the user wants one, but not the other.
There's no objective definition of evil software -- there is only whether it is wanted or not by the user. One very reasonable thing the user might want is to cede making that decision to someone who is better able to make that judgement. One very reasonable thing the user might want is to make that judgement for themselves.
Politifact has a good summary of how Social Security is not a Ponzi scheme that applies here: First, you know how the bank works (with respect to reserves, that is). The poker site was defrauding users by stated that they behaved one way and behaving in a different (more profitable) way. Second, banks are generally considered a sustainable business model, whereas the poker site's business model is distinctly unsustainable -- at some point, they will run out of money to pay off winners, even if there is no "run" on the site.
...seriously, what could break?
I don't have a big problem with the short release cycle, but you shouldn't ask this question. The answer is "more than you'd ever expect".
stray from the standards enough to break your corporate site?
Standards? Also, yes.
Both man and info can handle that reasonably well -- man does not require a one-to-one association between commands and man pages, and much documentation in man is better than "command line parameters". Organization does get a bit ugly in man, though.
and the pedantic arguments of Mega vs Megi
It's actually mega- vs. mebi-. Kibi, mebi, gibi, tebi, pebi, exbi.
Wow, that is the most obviously-bad curve fitting I have ever seen.
Well, second most. The most obviously bad graphed observed temperatures from 1998 to 2008 and showed a "trend" by drawing a line between the 1998 measurement and the 2008 measurement.
The ballpark observed and predicted warming rate is actually about 1 degree Celcius per century.
Experts don't claim four-figure accuracy on temperature prediction. They don't even claim accurate temperature prediction over short time scales, which is not surprising, given the difficulty of modeling short-term behaviors of chaotic systems.
Fortunately, science and the world at large probably don't give a shit about whether you will listen or your arbitrary requirements.