"Wild and Crazy": Interview with Palladium's Mario Juarez
R. A. Hettinga writes: > Juarez: You have the ability to establish the notion of trusted code which > can't be observed or modified. Moreover, information on your machine, which > is living in one of those vaults or one of the sub-vaults, or as storage on > your disc can be encrypted with machine specific secrets so that they are > functionally useless if they are stolen. [For example,] if the hard drive > gets pulled or copied. In other words, when the MB is fried because of some freak electrical surge, I'm screwed, because I can't put the HD into another machine and get the data off it? - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Ross's TCPA paper
though... > > I don't think that one should confuse contract limitations, or > limitations on enforceable contract limitations, with technological > limitations. There is nothing, for example, in any legal system that > forbids one from violating the law of gravity. > > One of the many problems with the use of the Digital Millenium > Copyright Act to enforce the technological control measures > in DVD's was that it was based on the rather weird theory > that it should be illegal to do something that someone > else tried, but failed, to make technologically impossible > to do. > > (Thus I am rather doubtful that Lessig's idea the everything is > code is useful for analytical, rather than rhetorical, purposes.) > > : I think lawyers will hate this. > > I don't see why we should. We don't hate the law of gravity > or the law of large numbers. > > -- > Peter D. Junger--Case Western Reserve University Law School--Cleveland, OH > EMAIL: [EMAIL PROTECTED]URL: http://samsara.law.cwru.edu > NOTE: [EMAIL PROTECTED] no longer exists > > - > The Cryptography Mailing List > Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED] -- -------- Paul A.S. Ward, Assistant Professor Email: [EMAIL PROTECTED] University of Waterloo [EMAIL PROTECTED] Department of Computer Engineering Tel: +1 (519) 888-4567 ext.3127 Waterloo, OntarioFax: +1 (519) 746-3077 Canada N2L 3G1 URL: http://shoshin.uwaterloo.ca/~pasward - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: biometrics
Bill Frantz writes: > > What would be really nice is to be able to have the same PIN/password for > everything. Do you really mean that? Sure, if I only have to remember one thing it is easier for me. It is also a complete nightmare if it is ever compromised. -- Paul A.S. Ward, Assistant Professor Email: [EMAIL PROTECTED] University of Waterloo [EMAIL PROTECTED] Department of Computer Engineering Tel: +1 (519) 888-4567 ext.3127 Waterloo, OntarioFax: +1 (519) 885-1208 Canada N2L 3G1 URL: http://shoshin.uwaterloo.ca/~pasward - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: CFP: PKI research workshop
Eric Rescorla writes: > <[EMAIL PROTECTED]> writes: > > > Eric Rescorla writes: > > > Ben Laurie <[EMAIL PROTECTED]> writes: > > > > And most (all?) commercial CAs then disclaim any responsibility for > > > > having actually checked that right correctly... > > > While this is true, I'd point out that all the security software > > > you're using disclaims any responsibility for not having gaping > > > security holes. > > > > If an automaker disclaimed liability for a vehicle, and a negligent > > design or manufacture resulted in injury or loss, it is my > > understanding that the liability disclaimer notwithstanding, the > > automaker would be held responsible. Why do we believe that the same > > would not be the case for software? > In that case, why should the liability also apply to CAs, despite their > disclaimers? Do you mean "why should," or "why shouldn't?" If the latter, then, sure, I believe it should. People running around in business selling products and services and then disclaiming any liability with regard to their performance _for_their_intended_task_ is, IMHO, wrong. Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: CFP: PKI research workshop
Eric Rescorla writes: > Ben Laurie <[EMAIL PROTECTED]> writes: > > > Michael Sierchio wrote: > > > > > > Carl Ellison wrote: > > > > > > > If that's not good enough for you, go to https://store.palm.com/ > > > > where you have an SSL secured page. SSL prevents a man in the middle > > > > attack, right? This means your credit card info goes to Palm > > > > Computing, right? Check the certificate. > > > > > > To be fair, most commercial CA's require evidence of "right to use" > > > a FQDN in an SSL server cert. But your point is apt. > > > > And most (all?) commercial CAs then disclaim any responsibility for > > having actually checked that right correctly... > While this is true, I'd point out that all the security software > you're using disclaims any responsibility for not having gaping > security holes. If an automaker disclaimed liability for a vehicle, and a negligent design or manufacture resulted in injury or loss, it is my understanding that the liability disclaimer notwithstanding, the automaker would be held responsible. Why do we believe that the same would not be the case for software? Paul Ward - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: private-sector keystroke logger...
Jay D. Dyson writes: > -BEGIN PGP SIGNED MESSAGE- > > On Tue, 27 Nov 2001 [EMAIL PROTECTED] wrote: > > > > > Hrm, how about a worm with a built-in HTTP server that installs itself > > > > on some non-standard port, say TCP/28462 (to pick one at random)? > > > > > > Craftier still, backdoor an existing service that behaves normally > > > until it receives a few specially-crafted packets, then it opens a high > > > port for direct login or data retrieval. > > > > Neither of these will get past a firewall on an uncompromised machine. > > While I didn't enumerate the service that could be backdoored, I > do believe Eric Murray hit the nail on the canonical head when he > mentioned that such a beastie could target the firewall's configuration, > forcing it to relax its stance enough to allow the automated intrusion > agent plenty of latitude to conduct its business. I am assuming a firewall on a separate machine, which simply does not allow incoming connections to the window's boxes, and constrains the outgoing connections. I do not claim that this prevents all covert loss of data, but it constrains the options, and certainly does not permit the described backdoor to work. Better still would be a firewall design that monitored user bahaviour, and so deviation from that behaviour could be detected. Again, not that this is perfect, but it further constrains the options of getting the data out. - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: IP: Magic Lantern
Adam Fields writes: > > On a somewhat related note, is it > wise for the FBI to open itself up to potential lawsuits if their > software corrupts data or otherwise interferes with legitimate > business, or allows an intruder to do so undetected by utilizing the > AV-invisibility channel reserved for FBI-ware (if such a thing > exists)? I can see it now: FBIware is now ready to complete installation of FBILogger [tm]. However, you must first read and agree to the following end-user licence agreement. This software is provided as is, with no warranty. Under no circumstance is FBIware responsible for loss or corruption of data. You may have additional rights according to the state you live in. On a slightly more serious note: given the multiplicity of software that has similar licence agreements, exactly how would you prove in court that it was the FBI's installed logger software that caused the problem? - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: private-sector keystroke logger...
Jay D. Dyson writes: > On 27 Nov 2001, Derek Atkins wrote: > > > Hrm, how about a worm with a built-in HTTP server that installs itself > > on some non-standard port, say TCP/28462 (to pick one at random)? > > Craftier still, backdoor an existing service that behaves normally > until it receives a few specially-crafted packets, then it opens a high > port for direct login or data retrieval. Neither of these will get past a firewall on an uncompromised machine. - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Jay D. Dyson writes: > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > But this doesn't really address the question. Certainly you take > > various precautions. The question is: how can I know if the system is > > compromised? > > There's a wealth of utilities that can indicate system compromise. > These tools range from Tripwire to the Advanced Intrusion Detection > Environment (AIDE), plus a range of network sniffing utilities that can be > configured to look for unusual traffic. There's also the CryptoFileSystem > that precludes the Great Forces of Malevolence from sneaking things onto > your drive without your knowledge. Thanks. > All of these security-enhancing features must be predicated by > cradle-to-grave security, though. That means trusted installation of a > trusted OS from a trusted source on a trusted, non-networked box. Coupled > with that is assured physical security of the system by tamper-evident > systems. I assume you mean non-networked at installation time, not afterwards. > In the final analysis, there's no substitute for simple human > vigilance and a healthy amount of paranoia. Not one of these tools are of > any use if you have a user at the helm who will gleefully download and > execute the latest trojan horse. I'm not entirely sure I believe that last statement. Let's say I have a tripwire-like system, but the process is constantly running. So you cannot compromise the code on disk in a useful fashion. What can a trojan actually do without being detected? - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Jay D. Dyson writes: > -BEGIN PGP SIGNED MESSAGE- > > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > > Yet another reason why Open Source is your friend. > > > > I did not mean to imply that I am running some variety of windows. I am > > interested in the technical problem of what is the state of the art for > > detecting whether or not a computer has been tampered with. The use of > > some version of un*x does not per se solve this. > > I'm afraid we're still in the "arms race" model in that respect. > Every time one party comes up with a new widget, another party quickly > follows with a widget-defeater. Then the original party releases an > updated widget with a widget-defeater-defeater feature. Then the opposing > party responds in kind. On and on it goes...like a dog chasing its tail. > > My original response handles the electronic portion of the > equation (though I do concede the point another writer made that all bets > are off when the day of the Backdoored BIOS arrives). If you mean only > the physical aspect of the equation, there are a number of tricks you can > use ranging from sealing a system with epoxy, locks and so on...or (for > those who dig Mission: Impossible stuff), boobytrapping a system to either > explode a dye-pack (like that used in banks) or commit digital seppuku if > an unauthorized party dicks with it. > > I must admit the dye-pack idea has a certain appeal to it. > Nothing would make my day like seeing some goons come out of my house with > bright Candy Apple Red faces. I'm not actually worried about physical access at this point. Breaking and entering is a lot more difficult that hacking into a system, and frequently leaves evidence. More to the point, this is no different as a risk than that experienced whenever you use a physical ATM machine to access cash. My concern is with software access to a machine that is to be used in the same manner as an ATM. Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Jay Sulzberger writes: > > > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > Jay Sulzberger writes: > > > > > > > > > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > > > > > R. A. Hettinga writes: > > > > > Everyone remember First Virtual's Nat Borenstein's "major discovery" of the > > > > > keyboard logger? > > > > > > > > > > 'Magic Lantern' part of new 'Enhanced Carnivore Project' > > > > > > > > > [etc] > > > > > > > > In the same vein, but a different application, does anyone know what > > > > the state of the art is for detecting such tampering? In particular, > > > > when sitting at a PC doing banking, is there any mechanism by which a > > > > user can know that the PC is not corrupted with such a key logger? > > > > The last time I checked, there was nothing other than the various > > > > anti-virus software. > > > > > > > > Paul > > > > > > If you are running a source secret operating system, it is more difficult > > > to detect tampering. > > > > I'm sure it is, unless you have to be the company that owns the > > "source-secret operating system," in which case you can presumably do > > whatever is done by an open-source system. Now, what (beyond AV and > > tripwire) is done? > > > > Paul > > There is much that the holder of copyright on a source secret OS could do. > But their best efforts would likely be less effective than the best > efforts called forth by the market forces which operate on free software. Unclear at this point. The fact that a certain company produces a poor OS, does not mean all secret source OSes are poor. Are AIX, HPUX, Solaris, VMS, VM, ... all worse than Linux on this point? They certainly tend to be tampered with far less. - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Jay Sulzberger writes: > > > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > R. A. Hettinga writes: > > > Everyone remember First Virtual's Nat Borenstein's "major discovery" of the > > > keyboard logger? > > > > > > 'Magic Lantern' part of new 'Enhanced Carnivore Project' > > > > > [etc] > > > > In the same vein, but a different application, does anyone know what > > the state of the art is for detecting such tampering? In particular, > > when sitting at a PC doing banking, is there any mechanism by which a > > user can know that the PC is not corrupted with such a key logger? > > The last time I checked, there was nothing other than the various > > anti-virus software. > > > > Paul > > If you are running a source secret operating system, it is more difficult > to detect tampering. I'm sure it is, unless you have to be the company that owns the "source-secret operating system," in which case you can presumably do whatever is done by an open-source system. Now, what (beyond AV and tripwire) is done? Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Jay D. Dyson writes: > On Wed, 21 Nov 2001 [EMAIL PROTECTED] wrote: > > > > Everyone remember First Virtual's Nat Borenstein's "major discovery" > > > of the keyboard logger? > > > > > > 'Magic Lantern' part of new 'Enhanced Carnivore Project' > > > > In the same vein, but a different application, does anyone know what the > > state of the art is for detecting such tampering? In particular, when > > sitting at a PC doing banking, is there any mechanism by which a user > > can know that the PC is not corrupted with such a key logger? The last > > time I checked, there was nothing other than the various anti-virus > > software. > > As much as this will sound like a panacean suggestion, I'd say the > best way to avoid being a victim of this sort of attack is to dump Windows > and utilize Linux (or Solaris x86) with a GUI front end. With the advance > of *nix GUIs and the advent of utility suites such as Sun Microsystems' > Star Office, I've long since abandoned any justification to continue using > the Microsoft Windows operating system and office-oriented applications. > > Yet another reason why Open Source is your friend. I did not mean to imply that I am running some variety of windows. I am interested in the technical problem of what is the state of the art for detecting whether or not a computer has been tampered with. The use of some version of un*x does not per se solve this. - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
Kent Borg writes: > On Wed, Nov 21, 2001 at 10:40:11AM -0500, [EMAIL PROTECTED] wrote: > > In the same vein, but a different application, does anyone know what > > the state of the art is for detecting such tampering? In particular, > > when sitting at a PC doing banking, is there any mechanism by which a > > user can know that the PC is not corrupted with such a key logger? > > The last time I checked, there was nothing other than the various > > anti-virus software. > > I can imagine an arms race between the Feds and anti-virus-types, that > is until the anti-virus programs are strong-armed one way or the other > into backing down. I am certain that will happen, either behind the > scenes or by public law. > > I think you are toast if you are sitting at a PC and the Feds ~really~ > want to catch your keystrokes. That is, if the Feds are acting > competently. They might be coy with their good keyloggers to keep > samizdat word of their details from getting out. They might save the > good stuff for important targets. My concern isn't with the Feds snooping. It is with some criminal who wants banking-type information so as to rob the account, though it would appear that solving the one implies solving the other. > Alternatively, to move to a physical analogy, instead of leaving a > telltale thread on your door and trying to spot intruders that way, > you might instead invest in good locks in the first place. That is, > to use a reasonably secure operating system. At risk of starting an > OS war, a well managed Linux box is going to be pretty secure. > > Or, for a practical example, I am typing this on a Linux notebook that > mostly is obscured behind firewalls. If I keep damn Javascript OFF > and don't launch viruses that might be sent to me, and don't reuse > passwords between here and an unsecure computer, I think they are > going to have a very hard time cracking in without my knowing. But this doesn't really address the question. Certainly you take various precautions. The question is: how can I know if the system is compromised? Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Shades of FV's Nathaniel Borenstein: Carnivore's "Magic Lantern"
R. A. Hettinga writes: > Everyone remember First Virtual's Nat Borenstein's "major discovery" of the > keyboard logger? > > 'Magic Lantern' part of new 'Enhanced Carnivore Project' > [etc] In the same vein, but a different application, does anyone know what the state of the art is for detecting such tampering? In particular, when sitting at a PC doing banking, is there any mechanism by which a user can know that the PC is not corrupted with such a key logger? The last time I checked, there was nothing other than the various anti-virus software. Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Rubber hose attack
P.J. Ponder writes: > The default settings will be the permanent settings for many users, and if > it is easier to buy something through a .Net affiliate than to shop > around, then the .Net sites will get a certain percentage of users just by > 'default'. They won't get all, certainly, but they will get some just > because of the path of least resistance. But it doesn't even matter whether or not you shop around to find a non-.Net provider. What matters is whether or not your credit, etc., information travels over the .Net system at some point. You have no way of knowing that! A year or so ago there was an article in comp.risks about a web site that some user felt was insecure, for whatever reason (though not with respect to the security of the communication). It gave the option of 'phoning in the order instead of using the browser. The problem was, the person at the other end of the 'phone simply entered the data into the web site. Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Rubber hose attack
Rick Smith at Secure Computing writes: > > While I would feel compassion for consumers > who are hurt or inconvenienced by some huge scam that exploited a poor > Microsoft security implementation, such a scenario would be entertaining to > watch. What makes you believe that you will not be that consumer? - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Re: Security Research (Was: Scarfo "keylogger", PGP )
David Jablon writes: > > Not until vendors are held legally accountable for negligent design. > > Maybe someday, somehow, there will be a class action law suit. > (I saw a recent infosec conference flyer that had some silly quote > about the annual cost of viruses or something being in the > $100,000,000,000 range. :-) This is probably a silly question, but why isn't such a class action lawsuit launched? The stock answer I always here is the EULA. However, it is my understanding that if a manufacturer (say a car company) tried to disclaim or limit liability in the manner in which the software industry does, any court would throw out the disclaimer and impose its own standard. Can you imagine buying a Ford Explorer with the statement like: "not liable for any damages Under no circumstances will our liability exceed the original cost of the product." ? Now, can the lawyers please correct my ignorance. Paul - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
Criminalizing crypto criticism
David Jablon writes: > > Yet, on a sad note, public crypto research has to stop. > One might think it could survive in purely academic circles. > But no, you'd have to be a fool to criticise even an academic paper. > Anybody, perhaps the resentful author, could co-opt the work for > Copy Protection, and off to jail you go. > > We seem to be entering the twilight zone -- the end of an exciting, > but brief era -- of public cryptography. There is life outside the USA. - The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
