Migrating from one address space to another is painful. Why make it more frequent by aiming low? Do you think migration would be any less painful in 14 years?
By my estimate, we have at least 40 years before this becomes necessary, as it would require a ~2^25 increase in the storage size of the systems it is aimed at before it would become viable. That's assuming we can continue building bigger disks for that long, and those pesky problems with the granularity of space-time are solved.
3) New applications: Broadband didn't just result in really fast web-page downloads - the entire online music industry stems from that. The original creators of TCP/IP had no idea that they were developing media on-demand, they were making it so that you could transfer bits from one archaic machine to another.
If you could suggest to me a new applications that needs over 8 billion times more storage capacity as top-of-the-range current systems, please, go ahead and introduce it. Just don't ask me for financing.
Building flexible, capable systems creates an environment where development isn't as constrained by limitations - resulting in new, unpredictable developments.
Solving the hardware problems is the first step, not building capabilities into a file system that cannot be used in the foreseeable future due to our lack of ability to manufacture hardware within 9 orders of magnitude as powerful as it would require.
And really; on the multi-terabyte systems this is designed for, an extra 8 bytes per pointer isn't a big deal.
64 bits can handle "multi-terabyte" perfectly adequately. You'd be looking at "multi-billion-terabyte" before you needed to upgrade to 128 bits.
We're somewhere in the region of 75 years away from needing to upgrade beyond 64 bits, assuming Moore's Law applies to disks and holds out for that long.
Assuming you're going to use 64 bit computers it takes just as long to compute 96 bits as 128 bits. If you're going to use more than 64 bits, the next step might as well be 128.
Computational speed isn't the factor here, it's data storage.
Say you have a 40Gb drive divided into 4K allocation units. The pointers in the inodes (or whatever equivalent tech they're using) to all of those blocks are going to take up approximately 640Mb of your disk space. 96 bits would drop this down to 480Mb. But you might as well stick with 64 bits, as a 64 bit filesystem could, as another poster pointed out, manage adequately on a filing system which was spread over every drive Seagate sold in the last year. All 18 million of them.
And I think he somehow underestimated the capacity, possibly by only allowing for 2^64 bytes on the disk, whereas with the system I describe above, you'd be able to have 2^78, or 4096 times as much data as he considered.
once you're going to expand past a 64-bit filesystem, there's not much point in going smaller than a 128-bit fileystem.
Why expand past a 64 bit filesystem. 64 bits with 1k blocks as your smallest addressable unit (which is more than reasonable for a filesystem this size) gives you 2^74 bytes to play with. For reference, that's 16 * 2^70 bytes = 16 * 2^30 terabytes, or "one hell of a lot of data".
I don't see why this is insightful. A 64 bit filesystem, with 1k block size can support about 16 billion terabytes of information.
At very high definition video rates, let's say you wanted 1512x1024 24 bit colour uncompressed, that's 4.5 Megabytes per frame; at 60 frames per second, that's 270 megabytes per second. A 16 billion terabyte file system could therefore store about 1 billion minutes, or somewhere in the region of 45,000 hours of footage.
Someone can't go to google, type "most efficient diesel engine" and hit "I'm Feeling Lucky".
Having no idea of the origin of that article, or the method by which the measurements it claims for the efficiency of the engine in question, I wouldn't necessarily trust it.
However, a little research into thermal engines has convinced me that I was wrong about the maximum efficiency, and that by having a suitably high temperature range available to the engine it might be possible to exceed 50%. The claims are at least credible, according to what I have read.
Alternative history is another genre. Sure, you could combine the alternative genre with the scifi genre, and the result is OK because they are both sub-genres of the fictional fantasy novel.
I've heard a lot of things, but science fiction/alternative history a sub-genre of fantasy?
Only a few days ago I was arguing with someone else that fantasy isn't a sub-genre of science fiction!
The three may be closely related, but none are a sub-genre of another.
Science fiction / alternative history crossovers often make very good stories, btw.
Since both parties clearly had vested interests in the work not being in the public domain, I suspect they neglected to mention the date of original publication. The judge therefore drew his conclusion (which seems to me, BTW, perfectly valid) based on the evidence presented to him (as he is required to do, I believe) without considering the possibility of the work being public domain.
Thermal junction noise (e.g. from a noisy diode) has a predictable distribution dependent on the temperature of the junction. All you need to do is measure the temperature with an additional circuit and use this reading to compensate for the distribution in order to get your standard uniform distributed numbers in the range 0.0 to 1.0.
Same reason why no company has build the perfect car that last forever. Could we, absolutely? Will corporate america allow it, hell no!
Much as you might wish they were, Americans are not the only people in the world who manufacture cars.
The Germans, particularly, try very hard to make cars that are very reliable. Have they succeeded in "build[ing] the perfect car that last(sic) forever?"
In fact the most efficient ICE is some diesel engine that's the size of a house and is over 50% efficient, if I properly recall.
I'm not sure you do. I recall reading somewhere that the laws of thermodynamics prevent a combustion engine reaching efficiencies greater than 50%. Although, again, I could be wrong about that too.
Given that such information will be withheld, allowing people to pay to get notice that some information regarding an unspecified vulnerability in a particular application three days before other people (along with the paying subscribers) get the detailed information doesn't seem to be an unethical practice.
What if one of the paying subscribers uses this information, which will almost certainly include the type and approximate location of the vulnerability, to start searching specifically for this vulnerability, and therefore manages to work out what it is and how to exploit it a couple of days before the patch is released/the rest of us know about it and can use any workarounds available?
I plugged it in and the first thing it did was install their security software *without asking me*.
Do they sell these things in the UK? Modifying the contents of a computer system in any way without authorisation (which may, admittedly, be implicit) is an offense here under the Computer Misuse Act 1984, I believe.
To give them a little (very little) slack... Some form of the password has to be stored away, so you can validate the user-input password. But this shouldn't be rocket science, since SSH, PGP, GPG, and even PasswordSafe have done exactly this type of thing for aeons. All Lexar has done is put it on a little piece of solid-state removable storage.
I don't know anything at all about PasswordSafe, but the other 3 don't store your passwords. They use them to encrypt a private key, and then when you enter a password, they decrypt the key and check to see if the decrypted key matches the public key that they know is supposed to correspond to it. No storage of password involved.
Actually Casimir Effect is negative pressure (due to lower levels of quantum fluctuations inside the cavity), which causes positive energy (it is theoretically possible to use it to perform work).
And the beatles broke up in 1970...so at the time the company appeared dead.
No, they have constantly been re-releasing back catalogue since that date. Just because a company has no _new_ products doesn't make them dead. They've been coasting along just fine, making a fairly sizable revenue in the background.
Actually, Apple Computers holds the trademark "Apple" applied to computer devices and software, because that's what they use when they sell them. You don't buy an Apple Computers iMac, you buy an Apple iMac.
Apple Corps holds the trademark to "Apple" applied to records, because that's what was printed on the labels of the records they published.
Now Apple Computers are selling records. Admittedly only for download over the internet, but it is still closer than I would accept, if I were in any way involved with Apple Corps. The solution is for Apple to stop using any of their Apple branding at any point in association with iTunes.
Trademark law doesn't exist only to prevent copying -- it exists to prevent any actions which may cause customers of one business to confuse the two.
Now, if I wanted to buy some Beatles music, and happened to know that they were published under the label with "apple" in its name, I might go to a search engine and search for "apple music". Guess what? iTunes comes up first. iTunes Music Store is the second link; I want to buy a record, so that's what I click. I've now been confused -- dumb consumer that I am, I've ended up purchasing the record through iTMS when that wasn't what I originally wanted to do. Or perhaps I haven't found what I wanted (I don't know if iTMS sells any Beatles tracks?)
This is _exactly_ the kind of situation trademark law is designed to prevent.
Presumably you can find and translate the system binaries to build a translated app, but wouldn't this constitute "reverse engineering" that most software licenses prohibit?
The right to reverse engineer software for the purposes of enabling interoperability (which this must qualify for) is legally protected in the EU and many other regions. A lot of us have the freedom to ignore that aspect of our EULAs.
When I can run Office 2003 natively inside Linux then we can talk.
You're posting on the wrong article. This software is only useful for running UNIX-ish applications, according to the linked web site. The article author must have got hold of the wrong end of the stick.
Slashdot Mirror
Migrating from one address space to another is painful. Why make it more frequent by aiming low? Do you think migration would be any less painful in 14 years?
By my estimate, we have at least 40 years before this becomes necessary, as it would require a ~2^25 increase in the storage size of the systems it is aimed at before it would become viable. That's assuming we can continue building bigger disks for that long, and those pesky problems with the granularity of space-time are solved.
3) New applications: Broadband didn't just result in really fast web-page downloads - the entire online music industry stems from that. The original creators of TCP/IP had no idea that they were developing media on-demand, they were making it so that you could transfer bits from one archaic machine to another.
If you could suggest to me a new applications that needs over 8 billion times more storage capacity as top-of-the-range current systems, please, go ahead and introduce it. Just don't ask me for financing.
Building flexible, capable systems creates an environment where development isn't as constrained by limitations - resulting in new, unpredictable developments.
Solving the hardware problems is the first step, not building capabilities into a file system that cannot be used in the foreseeable future due to our lack of ability to manufacture hardware within 9 orders of magnitude as powerful as it would require.
And really; on the multi-terabyte systems this is designed for, an extra 8 bytes per pointer isn't a big deal.
64 bits can handle "multi-terabyte" perfectly adequately. You'd be looking at "multi-billion-terabyte" before you needed to upgrade to 128 bits.
We're somewhere in the region of 75 years away from needing to upgrade beyond 64 bits, assuming Moore's Law applies to disks and holds out for that long.
Assuming you're going to use 64 bit computers it takes just as long to compute 96 bits as 128 bits. If you're going to use more than 64 bits, the next step might as well be 128.
Computational speed isn't the factor here, it's data storage.
Say you have a 40Gb drive divided into 4K allocation units. The pointers in the inodes (or whatever equivalent tech they're using) to all of those blocks are going to take up approximately 640Mb of your disk space. 96 bits would drop this down to 480Mb. But you might as well stick with 64 bits, as a 64 bit filesystem could, as another poster pointed out, manage adequately on a filing system which was spread over every drive Seagate sold in the last year. All 18 million of them.
And I think he somehow underestimated the capacity, possibly by only allowing for 2^64 bytes on the disk, whereas with the system I describe above, you'd be able to have 2^78, or 4096 times as much data as he considered.
once you're going to expand past a 64-bit filesystem, there's not much point in going smaller than a 128-bit fileystem.
Why expand past a 64 bit filesystem. 64 bits with 1k blocks as your smallest addressable unit (which is more than reasonable for a filesystem this size) gives you 2^74 bytes to play with. For reference, that's 16 * 2^70 bytes = 16 * 2^30 terabytes, or "one hell of a lot of data".
I don't see why this is insightful. A 64 bit filesystem, with 1k block size can support about 16 billion terabytes of information.
At very high definition video rates, let's say you wanted 1512x1024 24 bit colour uncompressed, that's 4.5 Megabytes per frame; at 60 frames per second, that's 270 megabytes per second. A 16 billion terabyte file system could therefore store about 1 billion minutes, or somewhere in the region of 45,000 hours of footage.
Get real.
Someone can't go to google, type "most efficient diesel engine" and hit "I'm Feeling Lucky".
Having no idea of the origin of that article, or the method by which the measurements it claims for the efficiency of the engine in question, I wouldn't necessarily trust it.
However, a little research into thermal engines has convinced me that I was wrong about the maximum efficiency, and that by having a suitably high temperature range available to the engine it might be possible to exceed 50%. The claims are at least credible, according to what I have read.
X has suid binaries.
Not necessarily
Alternative history is another genre. Sure, you could combine the alternative genre with the scifi genre, and the result is OK because they are both sub-genres of the fictional fantasy novel.
I've heard a lot of things, but science fiction/alternative history a sub-genre of fantasy?
Only a few days ago I was arguing with someone else that fantasy isn't a sub-genre of science fiction!
The three may be closely related, but none are a sub-genre of another.
Science fiction / alternative history crossovers often make very good stories, btw.
Star Wars was fantasy.
As long as you ignore midichlorians, which every right thinking geek should do.
Since both parties clearly had vested interests in the work not being in the public domain, I suspect they neglected to mention the date of original publication. The judge therefore drew his conclusion (which seems to me, BTW, perfectly valid) based on the evidence presented to him (as he is required to do, I believe) without considering the possibility of the work being public domain.
Who says you shouldn't run X on a server? Just make sure you have -nolisten tcp in the server setup. And for good measure, block the ports it uses.
Thermal junction noise (e.g. from a noisy diode) has a predictable distribution dependent on the temperature of the junction. All you need to do is measure the temperature with an additional circuit and use this reading to compensate for the distribution in order to get your standard uniform distributed numbers in the range 0.0 to 1.0.
Same reason why no company has build the perfect car that last forever. Could we, absolutely? Will corporate america allow it, hell no!
Much as you might wish they were, Americans are not the only people in the world who manufacture cars.
The Germans, particularly, try very hard to make cars that are very reliable. Have they succeeded in "build[ing] the perfect car that last(sic) forever?"
Err. No.
In fact the most efficient ICE is some diesel engine that's the size of a house and is over 50% efficient, if I properly recall.
I'm not sure you do. I recall reading somewhere that the laws of thermodynamics prevent a combustion engine reaching efficiencies greater than 50%. Although, again, I could be wrong about that too.
Given that such information will be withheld, allowing people to pay to get notice that some information regarding an unspecified vulnerability in a particular application three days before other people (along with the paying subscribers) get the detailed information doesn't seem to be an unethical practice.
What if one of the paying subscribers uses this information, which will almost certainly include the type and approximate location of the vulnerability, to start searching specifically for this vulnerability, and therefore manages to work out what it is and how to exploit it a couple of days before the patch is released/the rest of us know about it and can use any workarounds available?
I plugged it in and the first thing it did was install their security software *without asking me*.
Do they sell these things in the UK? Modifying the contents of a computer system in any way without authorisation (which may, admittedly, be implicit) is an offense here under the Computer Misuse Act 1984, I believe.
I use ROT-26.
I don't advise that. All crypto experts know you should do something unexpected to throw the analysts, like performing extra rounds or something.
I use 3 rounds of ROT-8 followed by one of ROT-2. They'll never work that one out.
To give them a little (very little) slack... Some form of the password has to be stored away, so you can validate the user-input password. But this shouldn't be rocket science, since SSH, PGP, GPG, and even PasswordSafe have done exactly this type of thing for aeons. All Lexar has done is put it on a little piece of solid-state removable storage.
I don't know anything at all about PasswordSafe, but the other 3 don't store your passwords. They use them to encrypt a private key, and then when you enter a password, they decrypt the key and check to see if the decrypted key matches the public key that they know is supposed to correspond to it. No storage of password involved.
It's probably stored while the device is using it to encrypt/decrypt data, and they forgot to zero it out afterwards.
Actually Casimir Effect is negative pressure (due to lower levels of quantum fluctuations inside the cavity), which causes positive energy (it is theoretically possible to use it to perform work).
And the beatles broke up in 1970...so at the time the company appeared dead.
No, they have constantly been re-releasing back catalogue since that date. Just because a company has no _new_ products doesn't make them dead. They've been coasting along just fine, making a fairly sizable revenue in the background.
Actually, Apple Computers holds the trademark "Apple" applied to computer devices and software, because that's what they use when they sell them. You don't buy an Apple Computers iMac, you buy an Apple iMac.
Apple Corps holds the trademark to "Apple" applied to records, because that's what was printed on the labels of the records they published.
Now Apple Computers are selling records. Admittedly only for download over the internet, but it is still closer than I would accept, if I were in any way involved with Apple Corps. The solution is for Apple to stop using any of their Apple branding at any point in association with iTunes.
Trademark law doesn't exist only to prevent copying -- it exists to prevent any actions which may cause customers of one business to confuse the two.
Now, if I wanted to buy some Beatles music, and happened to know that they were published under the label with "apple" in its name, I might go to a search engine and search for "apple music". Guess what? iTunes comes up first. iTunes Music Store is the second link; I want to buy a record, so that's what I click. I've now been confused -- dumb consumer that I am, I've ended up purchasing the record through iTMS when that wasn't what I originally wanted to do. Or perhaps I haven't found what I wanted (I don't know if iTMS sells any Beatles tracks?)
This is _exactly_ the kind of situation trademark law is designed to prevent.
Presumably you can find and translate the system binaries to build a translated app, but wouldn't this constitute "reverse engineering" that most software licenses prohibit?
The right to reverse engineer software for the purposes of enabling interoperability (which this must qualify for) is legally protected in the EU and many other regions. A lot of us have the freedom to ignore that aspect of our EULAs.
When I can run Office 2003 natively inside Linux then we can talk.
You're posting on the wrong article. This software is only useful for running UNIX-ish applications, according to the linked web site. The article author must have got hold of the wrong end of the stick.
I've run xnest inside xnest (repeat seven further times). OK, so its the same program, but does it count?