Ummm... I doubt that a copyright on a phrase would
stand up in court. A phrase wouldn't pass the test of being a significant or original work. And anyone
quoting a phrase from a larger work (which may be
legitimately copyrighted) could probably claim
exemption under the doctrine of "fair use".
But I agree that MPAA and RIAA are idiots for declaring war on P2P users.
Re:Cooling actually does speed up asynch CPUs
on
Clockless Computing
·
· Score: 1
It wouldn't be hard in this case. Nobody with
half a brain would believe that he believed
those statements were true.
Re:Cooling actually does speed up asynch CPUs
on
Clockless Computing
·
· Score: 1
One of the nice features of these chips is that they are tolerant of manufacturing errors. Often impurities in the silicon will change the resistance or capacitance of a long wire. In asynchronous designs, this just means operations that need that wire will be a little slower. In the synchronous world, either the whole chip fails or you have to underclock it.
Actually, that's a bit scary when you think about it. Suppose that I buy a computer with Ivanium III async microprocessor in it that is nominally rated at 2 Ghz. How would I know that the processor doesn't (say) execute floatingpoint
divide at 1/10th of nominal speed dues to a
manufacturing glitch? Can I get my money back?
Or what happens if (say) Boeing uses the Ivanium III in the flight control system of the 7a7. How does Boeing test the FCS processor boards? How can they be sure that a processor with a slow instruction caused by a manufacturing defect will not slip through their testing because said instruction is only executed when an engine stops in mid-flight.
(In reality Boeing would already be using redundant processors in safety critical situations like this. I'm just trying to highlight the potential difficulty of testing systems built using large-scale async processor chips.)
-- Steve
Re:Cooling actually does speed up asynch CPUs
on
Clockless Computing
·
· Score: 1
It's the classic change of stategy to ligitation to keep your company afloat.
I don't think that particular strategy should be a worry to an Open Source project. If the aim of litigation (or threat of litigation) is to raise money, nobody in their right mind would go after an asset-less Open Source project.
Other strategies are a worry of course; e.g. some company using a threat of patent litigation to hamstring a competing Open Source project. But even that has a risk for a company. Such a move would attract the ire of a lot of people, and (at the very least) generate a lot of bad publicity.
What about older windows machine when most people didn't have the net?
If your machine can't connect to the net, it typically doesn't matter much that its clock is off by a few minutes or minutes. (There are exceptions of course).
Skewed clocks are typically a problem when your computer talks to another computer in real time. For example, if you've mounted a remote file system on a machine with a skewed clock "make" can get confused by files that appear to have been modified in the future.
But will this ever happen? Probably not. There's too much money being made by the people who are in control of the "War on Drugs." Follow the money trail... you'll see that for the powerful, drug illegalization is profitable.
While there are clearly a lot of people making a lot of money (on both sides of the law) from illegal drugs, that's not the root problem, IMO.
The real problem is that it is politically expedient for politicians to keep drugs illegal, and make a big deal about fighting them. Being "hard on drugs" wins elections, and vice versa.
This is only going to change when the vast majority of the population come to the view that the cure is worse than the problem, and vote out the politicians who block change. I can't see this happening soon...
Here's how. Recall that an IBM 026/029 punch has a
duplicator reader on the left and a punch on the
right.
To make a straight duplicate a card, you put the card to be copied into the feed slot on the
reader, load a new card into the punch, then
hold down the duplicate toggle (I think). Easy.
Now suppose that you left out a character at
(say) column 50. Here's what you do. You put your card to be edited into the reader slot,
load a new card into the punch and then
carefully duplicate up to and including column
49. Next you put your thumb down HARD on the
card in the reader to stop it moving. Then
you type the character to be inserted. If you've done
it right, the card in the reader slip against the
wheel that advances is, and will still be at column 50. The punch will now be at column 51.
Finally, duplicate to the rest of the card.
To delete a character, you stop the the card
in the punch from moving while you hit the
"space" bar to move over the character you
want to delete.
Funny though... I can't recall an ERASE key on
the IBM 026/029 punches I used as an undergraduate.
Hey, I learned to program (and type!) using a punchcard machine. Can you remember the trick for inserting and deleting characters using an IBM 026 / 029 card punch?:-)
OK... so if I'm totally clueless, how come
the AV version of news report also apparently
said that the spy plane had a direct satellite
link? (I didn't download the AV myself...
this is what someone else reported.)
The real truth
about military comms capabilities and operational
comms utilisation is likely to be classified.
So how would you know? And why should we believe
you Mr Anonymous Coward?
But, even the "simple" case may not be as simple
as you think. It depends on how hard-wired their
comms centres are, and how inventive the tech
support people are. We can only speculate...
And don't forget the "organizational" issues. I can imagine conversations like this.
SigInt guy: I need to run crypto XYZ over this
unencrypted link.
Crypto guy: We cannot allow that.
SigInt guy: Why not??
Crypto guy: You don't have "need to know".
SigInt guy:...sigh... How do I solve my problem then?
Crypto guy:...click...
Maybe, maybe not. But even the best generals have been known to make short-sighted and down-right stupid decisions. Signals Intelligence people are no different. [Nor is anyone for that matter, military or civilian.]
The $1000 PC would need to be patched into (for example) the P-3's surveilance and comms circuits.
It is quite likely that this would require non-trivial bespoke hardware, not to mention things
like special power adapters, rack mounting, vibration and RF hardening and so on.
For an unmanned drone, it is likely to be that much harder
because of tight limits on available space and power.
Isn't the main problem that you generally cannot and don't want to get rid of users?
The traditional "solution" to high security is to
encase your computer in a concrete block and drop it into the nearest deep-sea trench. But I guess you could add your users to the (concrete) mix too. You could get rid of your users, and they can still use the system... for a short time:-)
Isn't the main problem that you generally cannot and don't want to get rid of users?
That depends on your system, and the nature of the attacks you are trying to guard against. Saying that users are the main problem is (IMO) an over-generalisation.
You cannot assume that the holes have been patched. Patching old versions of software is going to be a lower priority for software vendors. Indeed, it is common practice for vendors to "de-support" old versions. If you ask the vendor, they will start by saying "upgrade to the latest version". Indeed, they may even be unwilling (or unable) to provide a known patch that was developed before they de-supported the product.
Also note that people often uncover a security problem in the current version of some product that turns out to have been present in previous versions... including versions that have been de-supported. If this happens, a patch for your obsolete box may never be forthcoming.
Security depends on good system design and good programming and diligent systems administration and careful users. Throw in good physical security and reliable hardware for good measure. If any one of these links breaks down, your security could be gone.
"Have you disciplined your users today?" -- The System Adminastrix.
Slashdot Mirror
Verign is lucky that nobody sued their arses off ... at least morally, if not in law.
for negligence. Microsoft would have a strong
case
"The End of Hell as we know it predicted"
"One billion obligations fall due as Hell freezes over"
"Snowball gets new chance".
You are ignoring the cost to you in time and
...
raised blood pressure each time you read and
delete a SPAM SMS.
I'm glad I don't have a mobile phone
But I agree that MPAA and RIAA are idiots for declaring war on P2P users.
It wouldn't be hard in this case. Nobody with half a brain would believe that he believed those statements were true.
Actually, that's a bit scary when you think about it. Suppose that I buy a computer with Ivanium III async microprocessor in it that is nominally rated at 2 Ghz. How would I know that the processor doesn't (say) execute floatingpoint divide at 1/10th of nominal speed dues to a manufacturing glitch? Can I get my money back?
Or what happens if (say) Boeing uses the Ivanium III in the flight control system of the 7a7. How does Boeing test the FCS processor boards? How can they be sure that a processor with a slow instruction caused by a manufacturing defect will not slip through their testing because said instruction is only executed when an engine stops in mid-flight.
(In reality Boeing would already be using redundant processors in safety critical situations like this. I'm just trying to highlight the potential difficulty of testing systems built using large-scale async processor chips.) -- Steve
Look up "slander" in the dictionary Mr Critter92.
Yea, but if they have serious number of roubles, they can spare a few of them to fight off a corporate vampire.
I don't think that particular strategy should be a worry to an Open Source project. If the aim of litigation (or threat of litigation) is to raise money, nobody in their right mind would go after an asset-less Open Source project.
Other strategies are a worry of course; e.g. some company using a threat of patent litigation to hamstring a competing Open Source project. But even that has a risk for a company. Such a move would attract the ire of a lot of people, and (at the very least) generate a lot of bad publicity.
If your machine can't connect to the net, it typically doesn't matter much that its clock is off by a few minutes or minutes. (There are exceptions of course).
Skewed clocks are typically a problem when your computer talks to another computer in real time. For example, if you've mounted a remote file system on a machine with a skewed clock "make" can get confused by files that appear to have been modified in the future.
While there are clearly a lot of people making a lot of money (on both sides of the law) from illegal drugs, that's not the root problem, IMO.
The real problem is that it is politically expedient for politicians to keep drugs illegal, and make a big deal about fighting them. Being "hard on drugs" wins elections, and vice versa.
This is only going to change when the vast majority of the population come to the view that the cure is worse than the problem, and vote out the politicians who block change. I can't see this happening soon ...
At least they are not wiring it with coal (or Coke)
To make a straight duplicate a card, you put the card to be copied into the feed slot on the reader, load a new card into the punch, then hold down the duplicate toggle (I think). Easy.
Now suppose that you left out a character at (say) column 50. Here's what you do. You put your card to be edited into the reader slot, load a new card into the punch and then carefully duplicate up to and including column 49. Next you put your thumb down HARD on the card in the reader to stop it moving. Then you type the character to be inserted. If you've done it right, the card in the reader slip against the wheel that advances is, and will still be at column 50. The punch will now be at column 51. Finally, duplicate to the rest of the card.
To delete a character, you stop the the card in the punch from moving while you hit the "space" bar to move over the character you want to delete.
Funny though ... I can't recall an ERASE key on
the IBM 026/029 punches I used as an undergraduate.
Hey, I learned to program (and type!) using a punchcard machine. Can you remember the trick for inserting and deleting characters using an IBM 026 / 029 card punch? :-)
The real truth about military comms capabilities and operational comms utilisation is likely to be classified. So how would you know? And why should we believe you Mr Anonymous Coward?
Thank you for sharing your insight with us. Not to mention your dazzling command of the English language.
But, even the "simple" case may not be as simple as you think. It depends on how hard-wired their comms centres are, and how inventive the tech support people are. We can only speculate ...
And don't forget the "organizational" issues. I can imagine conversations like this.
SigInt guy: I need to run crypto XYZ over this unencrypted link. ...sigh... How do I solve my problem then?
...click...
Crypto guy: We cannot allow that.
SigInt guy: Why not??
Crypto guy: You don't have "need to know".
SigInt guy:
Crypto guy:
Maybe, maybe not. But even the best generals have been known to make short-sighted and down-right stupid decisions. Signals Intelligence people are no different. [Nor is anyone for that matter, military or civilian.]
The $1000 PC would need to be patched into (for example) the P-3's surveilance and comms circuits. It is quite likely that this would require non-trivial bespoke hardware, not to mention things like special power adapters, rack mounting, vibration and RF hardening and so on.
For an unmanned drone, it is likely to be that much harder because of tight limits on available space and power.
Witches have been using gravitomagic for years ... to power their broomsticks ... which is old (witches) hat.
:-)
The traditional "solution" to high security is to encase your computer in a concrete block and drop it into the nearest deep-sea trench. But I guess you could add your users to the (concrete) mix too. You could get rid of your users, and they can still use the system ... for a short time :-)
That depends on your system, and the nature of the attacks you are trying to guard against. Saying that users are the main problem is (IMO) an over-generalisation.
You cannot assume that the holes have been patched. Patching old versions of software is going to be a lower priority for software vendors. Indeed, it is common practice for vendors to "de-support" old versions. If you ask the vendor, they will start by saying "upgrade to the latest version". Indeed, they may even be unwilling (or unable) to provide a known patch that was developed before they de-supported the product.
Also note that people often uncover a security problem in the current version of some product that turns out to have been present in previous versions ... including versions that have been de-supported. If this happens, a patch for your obsolete box may never be forthcoming.
Security depends on good system design and good programming and diligent systems administration and careful users. Throw in good physical security and reliable hardware for good measure. If any one of these links breaks down, your security could be gone.
"Have you disciplined your users today?" -- The System Adminastrix.
I have always found Redhat unbearable, so how is this new? You guys made a typo, right?
:-)
I think you misspelt unwearable. HTH.