They "purchased" the LGPL JDBC driver and relicensed it as GPL. This ensures that physical linking will occur with their software (and thus the warning in the article about "circumventing" the drivers).
Unfortunately, they probably haven't thought about the fact that when a Java application uses JDBC it doesn't actually get linked to the JDBC driver. What happens is this:
- The application tells the Java system the name of the class that contains the driver. - The Java system loads the driver and links it to internal parts of Java. - The application uses internal parts of Java to access the database in a way that is entirely independent of the driver in use and is achieved through indirection.
I'd personally just distribute the Java application and tell the recipient you need a JDBC driver (any one will do) and database server. Oh, and you'll find MySQL and its JDBC driver on the same disk.
Yeah, I don't think SSH is affected. I believe it uses the entire public key of a connected entity to identify them (is this right? Someone here must know more about the protocol than I do). SSL, on the other hand, uses hashes (usually SHA-1) to verify the signature of the X509 certificates that are used to identify either client or server. If SHA-1 were completely broken, then an attacker would be able to imitate an SSL client or server.
If you can generate a collision with a known starting hash (which isn't what has been done here, incidentally), you can probably arrange it so that the algorithm produces a 'tail' to an initial data stream that produces the right hash. That's only a matter of changing the start state of the algorithm, which will usually be possible. In this case, you can produce a.tgz file that uncompresses to whatever you want and then fails as corrupted at the end -- but you might not notice that failure and just execute what decompresses successfully first. In the.exe case it's even easier... that 'tail' can be in a section of the file that isn't used at all, so no failure would occur.
one hash per second isn't being generous - a million hashes per second is.
That depends on the size of the message you're trying to generate. For most practical attacks (e.g. substituting one set of data for another) you'll be aiming at a message that's the same size as the original message. If that's up in the megabyte range, 1 second per hash is about right.
I haven't used a floppy disk all year. Mostly I use my iRiver iHP-120 as a portable storage device, when I can't just jack into the ethernet hub that all my friends have in their living rooms.
Can you boot off it? If you can, I'll accept it as a reasonable floppy substitute.
They DO however make "single" sized cd's that are rarely actually used for singles these days. They store up to 170MB i believe.
The main problem is that there are some CDROM drives that won't take them, primarily among the kind that load the disc through a slot in the front rather than having a tray you put the disc into.
after all, this 18-20 second delay you speak of, is just their inferior imitation of the 30-45 second wait most MS Office users experience...
Huh? I've never come across anyone who has a problem with how long MS Office takes to open. On my current workstation, a 450MHz celeron mounting the drive which office is installed on over a 100MB/s network, Word 97 takes 4 seconds to start. Including the time it takes for me to get rid of its annoying 'some files you need are missing' message box because I couldn't be bothered to install it locally.
By contrast, OO.o 1.1, installed on the local hard disk, takes 10 seconds to pop up its loading window, and a total of 27 seconds before the UI is usable.
Here's an idea: Performance will be nearly as good, reliability will be substantially up, cost will be a lot lower:
Use a traditional hard drive, but with a RAM cache that's as large as the drive. The drive controller uses idle time to preemptively load data into the cache. There's a battery backup so that the drive can continue operating after powerdown, and the system uses a long time period write behind cache with write combining to reduce drive usage in operation.
I've been getting that a lot lately. I think there's something wrong with the story posting mechanism so that it updates the front page, and then takes about 2-3 minutes to create the story pages.
a) Why would one transfer (copy!) any files when network bandwidth and performance would allow fast file access directly over network connection?
Think about it. In order to access something over a network you have to transfer it to your local working storage. That's the way such things work.
Besides, the context makes it sound like they're talking about some kind of "push" mechanism, which would generally need to upload content to be examined.
b) Why would one use "phone" when minds would be able to allow read-only access to things one'd like to communicate to each other
Privacy. Having to express something verbally allows you to filter the content to what you want to allow the other party to receive. Any form of direct mental link would probably decrease the amount of control available.
However, there is hope. The net is obviously revolutionizing publishing. These dinosaurs are in their very last days.
It is? Sorry, I don't see it. E-books might be taking off (slowly), but they're still published by companies working with the same traditional structure as the paper book world uses -- an editor reads incoming submissions, picks ones that look good, works with the author to improve them, and then they publish them.
While there are organisations that don't use this process, sales are typically tiny, except in rare cases where there is a strong 'word of mouth' movement over a particular book. I'll admit the advent of the blog has helped in some way here, but to be honest (a) the effect is tiny compared to the number of sales a professionally published book tends to get, and (b) there were plenty of pre-blog self published books that managed to do the same thing. I believe Tom Clancy's first novel was published in this way.
Sorry if I don't buy your 'net revolution thing, I just see no evidence for any change that has actually occurred.
But it's quantum computing that will likely allow us to solve NP problems; and a Quantum computer gets us no further towards AI.
Are you sure about that? One of the most promising AI approaches is the simulation of neural networks. Many people believe that intelligence will emerge once the networks reach sufficient complexity. However, simulating them is an NP problem...
Neural networks are very interesting things, and the hardware to experiment with them in real time (reconfigurable FPGAs and large computer memories) is just becoming available to low-budget and self-funded researchers. (yay!)
FPGAs are probably the wrong tool for the low budget researcher. I'd investigate modern 3D graphics hardware myself. I understand that the vertex shader pipeline on modern ATI cards can perform 750 million 4x4 matrix multiplications per second. The pixel shader pipelines run in parallel to this, and while I can't find a similar performance estimate for them the work they do is actually harder, so I would expect higher performance from them. These are programmable enough to support any kind of neural network you're likely to try to simulate. The cards cost about GBP280 (~ $400) from my local retailer, much cheaper than equivalent FPGAs.
It doesn't need fusion. If thinking at an amazing speed were important, then maybe.
I think "thinking at an amazing speed" is actually a fairly important part of what the singularity is about -- it's about machines (whether AI or augmented humans) that come up with new new ideas so rapidly that they completely change human culture. But you are right in one thing -- it isn't supposed to be difficult to achieve. In fact, if Vinge is right, it is almost unavoidable.
I think that the point, and the main question is, without a set physical vessel (emotions etc.), what would the point be?
I personally believe that emotions are a critical part of intelligence and that we're unlikely to ever produce or encounter an intelligence that does not have emotions (or at least some analogue to them). They are the control system that regulates behaviour in order to ensure the intelligence achieves productive things. In many senses, our emotions are trainers that supervise us to make sure we don't do anything stupid.
I thought the x-prize would push innovation forward, not recycling (or has the patent on v2 rockets recently expired) . Otherwise couldn't we have done this 50 years ago?
Given that the X prize is for private industry reproducing results that were first achieved about 50 years ago, I don't see it as surprising that they're recycling 50 year old technology. The point is, though, that it's a lot cheaper for these guys to do it now than it was then, because of improvements in other areas of technology (largely materials and manufacturing processes).
Also, note that the X prize vehicle must be reusable (2 launch requirement), which none of the technology of the 50s was, so they have to make some improvements.
Strange. For about three years after I moved into my first flat, I had no TV and, of course, no TV license. They never bothered me. No letters, no visits, nothing. The person who lived there before me had a license, so the address must have been on their records. Perhaps they concentrate their efforts on areas where they've found a lot of unlicensed TVs before?
They concentrate on addresses where one of the following has happened:
* They've had a report that the occupier has purchased a TV, but they have no record of a license * There previously was a license, but it expired and wasn't renewed.
The previous occupant of your flat probably changed the address on his license, and as you didn't buy a TV while there, they didn't bother you.
Brings back memories of the TV license people visiting university halls of residence. The cupboard on our hallway for mops and buckets had 12 TVs piled up in there as soon as the collectors appeared outside the front door:-)
My old university had a policy of not allowing them onto campus. As soon as anyone spotted them, they'd phone security who would turn up and escort them off the university's private land. That was a good arrangement.:)
Quick question for anyone who might be able to give me an answer here -- how do you set XP up to require a secure attention key sequence at login? I've tries setting the Local Security Policy setting "Interactive logon: Do not require CTRL+ALT+DEL" to "Disabled", but that had no effect, I can still log in by clicking on the user name in the list with the mouse and then typing the password. This is too easy to spoof with a fake login screen for my liking, and I would worry about deploying it on any general access machines.
This is a big problem because of the piss-poor privledge seperation in Windows.
Well, perhaps it is. But no other Windows firewall software (e.g. the one they are recommending) solves the problem.
no, "run as" does not count, since it doesn't work in the gui
I'm not sure what you mean by that. When I want to run a GUI application as another user, I create a shortcut to it, check the 'run as different user' box and then execute the shortcut. The only problem I have with MS's "run as" implementation is that it doesn't isolate applications from each other via DDE, so it's impossible to bring up an explorer window as another user (or any other app that checks for its existence in another process before starting up). But there are plenty of alternatives to using explorer -- I'll live.
Slashdot Mirror
They "purchased" the LGPL JDBC driver and relicensed it as GPL. This ensures that physical linking will occur with their software (and thus the warning in the article about "circumventing" the drivers).
Unfortunately, they probably haven't thought about the fact that when a Java application uses JDBC it doesn't actually get linked to the JDBC driver. What happens is this:
- The application tells the Java system the name of the class that contains the driver.
- The Java system loads the driver and links it to internal parts of Java.
- The application uses internal parts of Java to access the database in a way that is entirely independent of the driver in use and is achieved through indirection.
I'd personally just distribute the Java application and tell the recipient you need a JDBC driver (any one will do) and database server. Oh, and you'll find MySQL and its JDBC driver on the same disk.
Perfectly acceptable under the terms of the GPL.
Yeah, I don't think SSH is affected. I believe it uses the entire public key of a connected entity to identify them (is this right? Someone here must know more about the protocol than I do). SSL, on the other hand, uses hashes (usually SHA-1) to verify the signature of the X509 certificates that are used to identify either client or server. If SHA-1 were completely broken, then an attacker would be able to imitate an SSL client or server.
If you can generate a collision with a known starting hash (which isn't what has been done here, incidentally), you can probably arrange it so that the algorithm produces a 'tail' to an initial data stream that produces the right hash. That's only a matter of changing the start state of the algorithm, which will usually be possible. In this case, you can produce a .tgz file that uncompresses to whatever you want and then fails as corrupted at the end -- but you might not notice that failure and just execute what decompresses successfully first. In the .exe case it's even easier... that 'tail' can be in a section of the file that isn't used at all, so no failure would occur.
one hash per second isn't being generous - a million hashes per second is.
That depends on the size of the message you're trying to generate. For most practical attacks (e.g. substituting one set of data for another) you'll be aiming at a message that's the same size as the original message. If that's up in the megabyte range, 1 second per hash is about right.
Trying to find two inputs that produce the same output should ideally take at least 2^(N-1) steps.
Shouldn't that be 2^(N/2)?
I haven't used a floppy disk all year. Mostly I use my iRiver iHP-120 as a portable storage device, when I can't just jack into the ethernet hub that all my friends have in their living rooms.
Can you boot off it? If you can, I'll accept it as a reasonable floppy substitute.
They DO however make "single" sized cd's that are rarely actually used for singles these days. They store up to 170MB i believe.
The main problem is that there are some CDROM drives that won't take them, primarily among the kind that load the disc through a slot in the front rather than having a tray you put the disc into.
after all, this 18-20 second delay you speak of, is just their inferior imitation of the 30-45 second wait most MS Office users experience...
Huh? I've never come across anyone who has a problem with how long MS Office takes to open. On my current workstation, a 450MHz celeron mounting the drive which office is installed on over a 100MB/s network, Word 97 takes 4 seconds to start. Including the time it takes for me to get rid of its annoying 'some files you need are missing' message box because I couldn't be bothered to install it locally.
By contrast, OO.o 1.1, installed on the local hard disk, takes 10 seconds to pop up its loading window, and a total of 27 seconds before the UI is usable.
Here's an idea: Performance will be nearly as good, reliability will be substantially up, cost will be a lot lower:
Use a traditional hard drive, but with a RAM cache that's as large as the drive. The drive controller uses idle time to preemptively load data into the cache. There's a battery backup so that the drive can continue operating after powerdown, and the system uses a long time period write behind cache with write combining to reduce drive usage in operation.
its not designed for continuous disk activity such as the one which happens with swaps.
But, lets face it, you'd have to be dumb to set up a system that swaps onto a device that's more expensive per Gb than RAM.
I've been getting that a lot lately. I think there's something wrong with the story posting mechanism so that it updates the front page, and then takes about 2-3 minutes to create the story pages.
Within the decade the spinning hard disk may go the way of the floppy and CRT.
You mean it'll still be the default option on most new PCs and in use by ~90% of PC users?
This is the very same reason "No-CD" cracks are so damn popular.
;)
That, and the fact that they mean you can install a copy from a friend's CD.
I believe no such algorithm can be implemented because a Turing Machine by its nature cannot solve such problems when confronted with them.
OK, you're right. But how do you know that there aren't problems which have the same effect on humans?
a)
Why would one transfer (copy!) any files when network bandwidth and performance would allow fast file access directly over network connection?
Think about it. In order to access something over a network you have to transfer it to your local working storage. That's the way such things work.
Besides, the context makes it sound like they're talking about some kind of "push" mechanism, which would generally need to upload content to be examined.
b)
Why would one use "phone" when minds would be able to allow read-only access to things one'd like to communicate to each other
Privacy. Having to express something verbally allows you to filter the content to what you want to allow the other party to receive. Any form of direct mental link would probably decrease the amount of control available.
However, there is hope. The net is obviously revolutionizing publishing. These dinosaurs are in their very last days.
It is? Sorry, I don't see it. E-books might be taking off (slowly), but they're still published by companies working with the same traditional structure as the paper book world uses -- an editor reads incoming submissions, picks ones that look good, works with the author to improve them, and then they publish them.
While there are organisations that don't use this process, sales are typically tiny, except in rare cases where there is a strong 'word of mouth' movement over a particular book. I'll admit the advent of the blog has helped in some way here, but to be honest (a) the effect is tiny compared to the number of sales a professionally published book tends to get, and (b) there were plenty of pre-blog self published books that managed to do the same thing. I believe Tom Clancy's first novel was published in this way.
Sorry if I don't buy your 'net revolution thing, I just see no evidence for any change that has actually occurred.
Which has... what relevance to this story? None, as far as I can see.
But it's quantum computing that will likely allow us to solve NP problems; and a Quantum computer gets us no further towards AI.
Are you sure about that? One of the most promising AI approaches is the simulation of neural networks. Many people believe that intelligence will emerge once the networks reach sufficient complexity. However, simulating them is an NP problem...
Neural networks are very interesting things, and the hardware to experiment with them in real time (reconfigurable FPGAs and large computer memories) is just becoming available to low-budget and self-funded researchers. (yay!)
FPGAs are probably the wrong tool for the low budget researcher. I'd investigate modern 3D graphics hardware myself. I understand that the vertex shader pipeline on modern ATI cards can perform 750 million 4x4 matrix multiplications per second. The pixel shader pipelines run in parallel to this, and while I can't find a similar performance estimate for them the work they do is actually harder, so I would expect higher performance from them. These are programmable enough to support any kind of neural network you're likely to try to simulate. The cards cost about GBP280 (~ $400) from my local retailer, much cheaper than equivalent FPGAs.
It doesn't need fusion. If thinking at an amazing speed were important, then maybe.
I think "thinking at an amazing speed" is actually a fairly important part of what the singularity is about -- it's about machines (whether AI or augmented humans) that come up with new new ideas so rapidly that they completely change human culture. But you are right in one thing -- it isn't supposed to be difficult to achieve. In fact, if Vinge is right, it is almost unavoidable.
I think that the point, and the main question is, without a set physical vessel (emotions etc.), what would the point be?
I personally believe that emotions are a critical part of intelligence and that we're unlikely to ever produce or encounter an intelligence that does not have emotions (or at least some analogue to them). They are the control system that regulates behaviour in order to ensure the intelligence achieves productive things. In many senses, our emotions are trainers that supervise us to make sure we don't do anything stupid.
I thought the x-prize would push innovation forward, not recycling (or has the patent on v2 rockets recently expired) . Otherwise couldn't we have done this 50 years ago?
Given that the X prize is for private industry reproducing results that were first achieved about 50 years ago, I don't see it as surprising that they're recycling 50 year old technology. The point is, though, that it's a lot cheaper for these guys to do it now than it was then, because of improvements in other areas of technology (largely materials and manufacturing processes).
Also, note that the X prize vehicle must be reusable (2 launch requirement), which none of the technology of the 50s was, so they have to make some improvements.
Strange. For about three years after I moved into my first flat, I had no TV and, of course, no TV license. They never bothered me. No letters, no visits, nothing. The person who lived there before me had a license, so the address must have been on their records. Perhaps they concentrate their efforts on areas where they've found a lot of unlicensed TVs before?
They concentrate on addresses where one of the following has happened:
* They've had a report that the occupier has purchased a TV, but they have no record of a license
* There previously was a license, but it expired and wasn't renewed.
The previous occupant of your flat probably changed the address on his license, and as you didn't buy a TV while there, they didn't bother you.
Brings back memories of the TV license people visiting university halls of residence. The cupboard on our hallway for mops and buckets had 12 TVs piled up in there as soon as the collectors appeared outside the front door :-)
:)
My old university had a policy of not allowing them onto campus. As soon as anyone spotted them, they'd phone security who would turn up and escort them off the university's private land. That was a good arrangement.
Quick question for anyone who might be able to give me an answer here -- how do you set XP up to require a secure attention key sequence at login? I've tries setting the Local Security Policy setting "Interactive logon: Do not require CTRL+ALT+DEL" to "Disabled", but that had no effect, I can still log in by clicking on the user name in the list with the mouse and then typing the password. This is too easy to spoof with a fake login screen for my liking, and I would worry about deploying it on any general access machines.
This is a big problem because of the piss-poor privledge seperation in Windows.
Well, perhaps it is. But no other Windows firewall software (e.g. the one they are recommending) solves the problem.
no, "run as" does not count, since it doesn't work in the gui
I'm not sure what you mean by that. When I want to run a GUI application as another user, I create a shortcut to it, check the 'run as different user' box and then execute the shortcut. The only problem I have with MS's "run as" implementation is that it doesn't isolate applications from each other via DDE, so it's impossible to bring up an explorer window as another user (or any other app that checks for its existence in another process before starting up). But there are plenty of alternatives to using explorer -- I'll live.
and its for an account other than your own
I don't understand that, sorry. What do you mean?