... it's TPM glue for hard drives. The spec says almost nothing about encryption and authentication, it's just a bunch of TPM command and control mechanisms for hard drives. The IEEE P1696 working group is the one working on secure hard-drive encryption. Unfortunately the TPM people have better PR people than the CS and EE types doing the IEEE work do.
"IPAT, which apparently purchased these patents from their listed inventor of Addison M. Fischer"
Addison Fischer is sort of the man behind the men in a range of security-related make-money-from-IP deals. For example if you'd scratched the surface of RSA Data Security about 10 years ago when they still held a monopoly patent on the algorithm you'd have found him there somewhere, although you'd have to scratch pretty deep since he doesn't seem to like publicity much (he's an ex-spook, which may explain it).
Is this widely implemented? I've never heard of it before, although I'm by no means claiming any expertise. How does one go about setting it up?
Most servers and browsers support it, but any browser that connects to a server doing this is going to flash up all manner of warnings of doom and destruction because you haven't paid the appropriate CA tax to be allowed to run your server.
Yet it's been in use with Japanese banks for years. You'd think with such poor performance they'd have abandoned it, which makes me think that what the NPL studied and what Hitachi are selling might not be exactly the same thing.
The problem with any security system like this is that poor performance doesn't mean obviously poor performance. Take a biometrics matching system and wind the false acceptance rate up so that you have a false rejection rate of zero and it'll perform just brilliantly: everyone gets in without any problems (including lots of people who shouldn't, but the vendors never test for that). It's like a firewall with "allow from any to any", it works wonderfully and all the users are happy, pity it's not doing anything useful. I think that'd be the difference between the Hitachi and NPL results, one evaluated the system as it was theorized and advertised and the other as it was manufactured and employed.
Here's how to defeat something like this. The way the sensors work is they use IR to image your vein patterns which then show up as a monochrome image of strong dark lines on a pale background, just the sort of thing a basic cheap laser printer will reproduce perfectly. Put your lased/photocopied image on the sensor, press some body part onto it to fool the liveness sensors (if they're even using any, many don't bother), and you're in. You don't even need anything as complex as gummy fingers, just a desktop printer will do.
Is there a place to check this? I found the article you're talking about - but you have to buy it to read it.
I saw it as a paper copy so I'm not sure where you'd get it. I seem to recall that Ross Anderson mentions it in the second edition of "Security Engineering", which Google Books has here.
An evaluation by the National Physical Laboratory in the UK found vein patterns to be the least reliable biometric they'd ever encountered, worse even than face recognition which became notorious for its zero-percent hit rate in several public trials (OK, so you can't get worse than zero percent, but in carefully controlled lab trials face recognition did get a non-zero score).
Looks like another great example of biometric vendor marketing at work. "Buy our stuff, it's gooder than anyone else's!".
A big problem with this is that the definition of "dual-use technologies" covers almost anything more complex than an electric blanket. Some years ago as part of a due diligence exercise I had to perform a check against the export-control lists and used a Dell flyer that had arrived that day as a test case. Almost every single product on this generic Dell sales brochure violated at least one and sometimes several restrictions in the export control lists: chips with more than 208 pins, graphics performance above some early-90s level, software that performed network routing, the number of pitfalls is endless. So waving your hands and making a fuss about "illegal technology exports" sounds pretty scary until you realise they're talking about things like a 486 laptop with a Trio64 video card running Slackware 3.1.
Hmmm...I would have thought that QKDS have some way of checking that the information was received properly. For example before you distribute the key. You encrypt a short message. Then you send it to the recipient to decrypt. If the key was intercepted then the message would be unreadable.
That assumes that the system is working as the vendor claims. If I give you a QKD box that communicates the key by flashing an LED down the link then the crypto will pass your test but be totally insecure. I'm using "flashing an LED down the link" to make a point, any subtle failure in the QKD system can make it appear to function correctly but be insecure, and there's no way you can tell. OTOH with any random OSS equivalent (e.g. OpenVPN) you can run a self-test at any time to verify that the crypto is doing what it's supposed to.
It's not so much Windows instant-on that people care about, it's something better than the same-day service currently offered by Vista after you hit the power switch. If you could tune the OS to boot in 10-20s (or even 5s like the Linux Plumbers did recently this wouldn't even be something you'd need to ask users about.
Makes for a great news media sound bite, but what they've done is implemented Idea #2 of the
Six Dumbest Ideas in Computer Security. Still, as long as it drives up their public visibility and stock price, who cares whether it works or not.
A quantum key distribution system is NOT unbreakable encryption. Period. It simply gives you perfect assurance that your encryption keys are given (and hopefully known) only to the person they are intended.
Actually it doesn't even give you that. The whole point of QC is that you can't intercept the communication without destroying it. This means that you have no way of verifying whether your "security" is actually working - you could be doing the QC equivalent of rot26 but there's no way to check it, you just have to hope it's working OK. This is why security standards like FIPS 140 have extensive requirements for in-depth self-tests, and high-end crypto units run two lots of crypto in parallel with one cross-checking the other. With QC the only checking you have is "the vendor claims it's working as intended".
You've obviously not looked at much virus, worm, or malware software. It's mostly crap, assembled by people who think that inventing their own version of a sorting function or a password checker makes them 3l33t. Some of it is insightful, but mostly it's assembled like kids building go-carts from a junkyard of parts.
Ten years ago this was certainly true. A lot of the commercial malware coming out of Russia today is as well written or better written (and certainly better-tested!) than standard commercial software. In capitalist Russia........ Profit!
I'm of the school that thinks "debunking" involves some kind of comprehensive stats or numbers or evidence weight against strongly held opinions.
This article is basically a verbose version of the "nuh uh" argument.
Pretty much, and some of their claims are rather dubious, or based on really odd ideas (NiCd batteries in laptops? They haven't been used for that in 20 years!). Their point #1 just seems like nonsense, if they're going to make a claim like that they better have some pretty heavy evidence to back it up. Point #5 is at best a distortion, there are a range of monitor power mgt.schemes that may be enabled, and the PC may have it enabled by default as well. Point #7 is a weird mangling of information on LiIon batteries that only just corresponds to real life, and isn't actually useful for anyone wanting help in prolonging battery life. Point #9 is just as bad, they've mangled a complex situation into a single paragraph that fails to provide any useful information.
Also invariably the levels of security are something like:
1. User must know how to flip a power switch. 2. User must know how to plug in a network cable. 3. [Whatever it was that was circumvented in the story]. 4. ??? 5. Profit!
Sumatra has a quirky user interface that makes it rather a pain to use. I've also never really seen any evidence that its rendering quality is better than Foxit or Acrobat or whatever. It is nice and quick though.
Governments (well some governments anyway) seem to be outstandingly good at providing comprehensible, sensible privacy policies. Look at the one from the Australian Institute of Criminology for example or the New Zealand Police (those sites chosen because they're organisations that some people would be a bit nervous about:-). They tell you exactly what they collect, how they collect it, why they collect it, what they do with it, and how to disable some of it (e.g. cookies) if it makes you feel uncomfortable.
Now compare it to Telsta's policy which more or less says "We'll do anything we feel like with your personal data" - is there anyone in Australia that isn't included in some manner in their list of organisations that they'll hand your details to?.
So we need some standardisation for EULAs, just like foods must list their ingredients in some standard way.
Analyze the available EULAs, 90% of it boils probably down to the same few terms.
The major problem with CGI pron is that Uncanny Valley can take on a whole new meaning. If they stuck to modelling the physics of silicone instead of real flesh, that problem would be quickly solved.
NoScript is designed so the user can protect themselves. SSP is designed so that the website owner can protect users from other users, malicious widget developers, or perhaps unscrupulous CDNs. So you have to trust the site that's being used to attack you not to attack you? Isn't this a variation of asking the drunk if he's drunk?
I don't know how Debian's actions compare to other distributions. Debian seems to be particularly bad in this regard, vastly more so than any other distro. Especially annoying are the gratuitous changes they make based on obscure Debian religious dogma that you need to be smoking three different types of crack to grasp, they're changes made not for any good technical reason but because "our religion dictates that we do it this way and not that way".
And if you read the thread, he was given the right answer: build with -DPURIFY and all problems will be solved. It's not quite that simple, he was using Valgrind but the response was to use a Purify-specific tweak to the build. It's a bit like reporting a bug with gcc and being told a Visual C++ option to change in response.
Who are 'they' and 'them' and why are you assuming you know what they need? Seems like ASUS Eee's success suggests that it nailed what 'they' and 'them' need.
You've pretty much nailed why every so-called EEE-killer has failed dismally on the market: the vendors' complete lack of understanding of why the EEE is so popular. All the competitors decide to take the EEE and upgrade the CPU, add more memory, a hard drive, a larger screen, a larger keyboard... the result is yet another cheap crappy generic laptop that weighs 2-3 times as much as the EEE, is twice the size, and costs a lot more. The reason why Asus can't keep up with production of the EEE due to its popularity is because it's tiny, light, and cheap. Anyone who fails to understand that (as pretty much every other manufacturer has so far) is going to fail in creating an EEE-killer.
And as for the spate of recent news stories of EEE-killers, it's just a rehash of the endless stories of iPod-killers of previous years.
Slashdot Mirror
... it's TPM glue for hard drives. The spec says almost nothing about encryption and authentication, it's just a bunch of TPM command and control mechanisms for hard drives. The IEEE P1696 working group is the one working on secure hard-drive encryption. Unfortunately the TPM people have better PR people than the CS and EE types doing the IEEE work do.
I believe that you and the AC you replied to, need to get a sense of perspective
I agree completely. Let's stick Darl in the Total Perspective Vortex.
"IPAT, which apparently purchased these patents from their listed inventor of Addison M. Fischer"
Addison Fischer is sort of the man behind the men in a range of security-related make-money-from-IP deals. For example if you'd scratched the surface of RSA Data Security about 10 years ago when they still held a monopoly patent on the algorithm you'd have found him there somewhere, although you'd have to scratch pretty deep since he doesn't seem to like publicity much (he's an ex-spook, which may explain it).
Most servers and browsers support it, but any browser that connects to a server doing this is going to flash up all manner of warnings of doom and destruction because you haven't paid the appropriate CA tax to be allowed to run your server.
The problem with any security system like this is that poor performance doesn't mean obviously poor performance. Take a biometrics matching system and wind the false acceptance rate up so that you have a false rejection rate of zero and it'll perform just brilliantly: everyone gets in without any problems (including lots of people who shouldn't, but the vendors never test for that). It's like a firewall with "allow from any to any", it works wonderfully and all the users are happy, pity it's not doing anything useful. I think that'd be the difference between the Hitachi and NPL results, one evaluated the system as it was theorized and advertised and the other as it was manufactured and employed.
Here's how to defeat something like this. The way the sensors work is they use IR to image your vein patterns which then show up as a monochrome image of strong dark lines on a pale background, just the sort of thing a basic cheap laser printer will reproduce perfectly. Put your lased/photocopied image on the sensor, press some body part onto it to fool the liveness sensors (if they're even using any, many don't bother), and you're in. You don't even need anything as complex as gummy fingers, just a desktop printer will do.
I saw it as a paper copy so I'm not sure where you'd get it. I seem to recall that Ross Anderson mentions it in the second edition of "Security Engineering", which Google Books has here.
An evaluation by the National Physical Laboratory in the UK found vein patterns to be the least reliable biometric they'd ever encountered, worse even than face recognition which became notorious for its zero-percent hit rate in several public trials (OK, so you can't get worse than zero percent, but in carefully controlled lab trials face recognition did get a non-zero score).
Looks like another great example of biometric vendor marketing at work. "Buy our stuff, it's gooder than anyone else's!".
A big problem with this is that the definition of "dual-use technologies" covers almost anything more complex than an electric blanket. Some years ago as part of a due diligence exercise I had to perform a check against the export-control lists and used a Dell flyer that had arrived that day as a test case. Almost every single product on this generic Dell sales brochure violated at least one and sometimes several restrictions in the export control lists: chips with more than 208 pins, graphics performance above some early-90s level, software that performed network routing, the number of pitfalls is endless. So waving your hands and making a fuss about "illegal technology exports" sounds pretty scary until you realise they're talking about things like a 486 laptop with a Trio64 video card running Slackware 3.1.
Hmmm...I would have thought that QKDS have some way of checking that the information was received properly. For example before you distribute the key. You encrypt a short message. Then you send it to the recipient to decrypt. If the key was intercepted then the message would be unreadable.
That assumes that the system is working as the vendor claims. If I give you a QKD box that communicates the key by flashing an LED down the link then the crypto will pass your test but be totally insecure. I'm using "flashing an LED down the link" to make a point, any subtle failure in the QKD system can make it appear to function correctly but be insecure, and there's no way you can tell. OTOH with any random OSS equivalent (e.g. OpenVPN) you can run a self-test at any time to verify that the crypto is doing what it's supposed to.
It's not so much Windows instant-on that people care about, it's something better than the same-day service currently offered by Vista after you hit the power switch. If you could tune the OS to boot in 10-20s (or even 5s like the Linux Plumbers did recently this wouldn't even be something you'd need to ask users about.
Makes for a great news media sound bite, but what they've done is implemented Idea #2 of the Six Dumbest Ideas in Computer Security. Still, as long as it drives up their public visibility and stock price, who cares whether it works or not.
Actually it doesn't even give you that. The whole point of QC is that you can't intercept the communication without destroying it. This means that you have no way of verifying whether your "security" is actually working - you could be doing the QC equivalent of rot26 but there's no way to check it, you just have to hope it's working OK. This is why security standards like FIPS 140 have extensive requirements for in-depth self-tests, and high-end crypto units run two lots of crypto in parallel with one cross-checking the other. With QC the only checking you have is "the vendor claims it's working as intended".
In game theory this is sometimes referred to as the "asking the drunk whether he's drunk" strategy. It works about as well as could be expected.
You've obviously not looked at much virus, worm, or malware software. It's mostly crap, assembled by people who think that inventing their own version of a sorting function or a password checker makes them 3l33t. Some of it is insightful, but mostly it's assembled like kids building go-carts from a junkyard of parts.
Ten years ago this was certainly true. A lot of the commercial malware coming out of Russia today is as well written or better written (and certainly better-tested!) than standard commercial software. In capitalist Russia........ Profit!
I'm of the school that thinks "debunking" involves some kind of comprehensive stats or numbers or evidence weight against strongly held opinions.
This article is basically a verbose version of the "nuh uh" argument.
Pretty much, and some of their claims are rather dubious, or based on really odd ideas (NiCd batteries in laptops? They haven't been used for that in 20 years!). Their point #1 just seems like nonsense, if they're going to make a claim like that they better have some pretty heavy evidence to back it up. Point #5 is at best a distortion, there are a range of monitor power mgt.schemes that may be enabled, and the PC may have it enabled by default as well. Point #7 is a weird mangling of information on LiIon batteries that only just corresponds to real life, and isn't actually useful for anyone wanting help in prolonging battery life. Point #9 is just as bad, they've mangled a complex situation into a single paragraph that fails to provide any useful information.
Also invariably the levels of security are something like:
1. User must know how to flip a power switch.
2. User must know how to plug in a network cable.
3. [Whatever it was that was circumvented in the story].
4. ???
5. Profit!
Sumatra has a quirky user interface that makes it rather a pain to use. I've also never really seen any evidence that its rendering quality is better than Foxit or Acrobat or whatever. It is nice and quick though.
Governments (well some governments anyway) seem to be outstandingly good at providing comprehensible, sensible privacy policies. Look at the one from the Australian Institute of Criminology for example or the New Zealand Police (those sites chosen because they're organisations that some people would be a bit nervous about :-). They tell you exactly what they collect, how they collect it, why they collect it, what they do with it, and how to disable some of it (e.g. cookies) if it makes you feel uncomfortable.
Now compare it to Telsta's policy which more or less says "We'll do anything we feel like with your personal data" - is there anyone in Australia that isn't included in some manner in their list of organisations that they'll hand your details to?.
This is why we have the EULAyzer.
Has Picasa finally got to the point where it won't crash any more if you're not Administrator when you run it under Windows?
You've pretty much nailed why every so-called EEE-killer has failed dismally on the market: the vendors' complete lack of understanding of why the EEE is so popular. All the competitors decide to take the EEE and upgrade the CPU, add more memory, a hard drive, a larger screen, a larger keyboard... the result is yet another cheap crappy generic laptop that weighs 2-3 times as much as the EEE, is twice the size, and costs a lot more. The reason why Asus can't keep up with production of the EEE due to its popularity is because it's tiny, light, and cheap. Anyone who fails to understand that (as pretty much every other manufacturer has so far) is going to fail in creating an EEE-killer.
And as for the spate of recent news stories of EEE-killers, it's just a rehash of the endless stories of iPod-killers of previous years.