December 2003 - cypherpunks-legacy

Re: "Key Escrow" --- the very idea
by Carl Ellison 17 Dec '03

17 Dec '03

if you really want to propose an escrow system we can live with, I would demand that it include: 1. unambiguous ID of the person being tapped in the LEAF-equivalent 2. multiple escrow agencies, at least one of which is the NSA HQ (for its superior physical security) 3. watchdogs as escrow agents (e.g., ACLU, Rep & Dem parties, CPSR, EFF, NYTimes, ...) with authorization to look for abuses of authority and to refuse to release keys in such cases and to publicize such cases as well as bringing them to the attention of law enforcement for prosecution. 4. user-generated escrow keys, to reduce the chance of anyone having a backdoor way to get the whole escrow key database.

9 10

(None)
by Jeff Gostin 17 Dec '03

17 Dec '03

adwestro(a)ouray.Denver.Colorado.EDU (Alan Westrope) writes: > Sorry, but there is NO escrow system I can live with -- I don't > care if John Gilmore is selected to head the escrow agency. Here Here! I don't care if I'M selected to head the escrow agency. I don't (or wouldn't) trust any of my underlings with my security. --Jeff -- ====== ====== +----------------jgostin(a)eternal.pha.pa.us----------------+ == == | The new, improved, environmentally safe, bigger, better,| == == -= | faster, hypo-allergenic, AND politically correct .sig. | ==== ====== | Now with a new fresh lemon scent! | PGP Key Available +---------------------------------------------------------+

1 0

by the way...
by perry＠imsi.com 17 Dec '03

17 Dec '03

For those who believe "the NSA can do ANYTHING" or some such, an article in the New York Times claims the annual black budget now seems to be hovering around $28 Billion per year, for ALL secret government work. All the spy planes, CIA bribes, etc, come out of that pool. Its a lot of money, but not enough to pay for really outlandish things, like disk drives the size of Seattle to store exhaustive listings of all possible DES encryptions. Perry

3 2

Re: "Key Escrow" --- the very idea
by Berzerk 17 Dec '03

17 Dec '03

On Fri, 22 Jul 1994 Mike_Spreitzer.PARC(a)xerox.com wrote: > Eight pieces seems too few to me. It's too easy for gov't agencies to "lean > on" eight individuals or organizations (someone else suggested "watchdog" > groups as fragment holding agencies, but that doesn't seem very good. Groups > can change over time, respond to pressure. Putting a lot of fragments in a few > hands seems fairly fundamentally flawed). I'd rather see thousands. That way, NO, what you really need to do is tackle the issue of the government rounding up keys in mass, and instituting an orwellian system of spying. To do this, simply make it legal for the escrow agencies to distroy their database as a whole, in fact, make it a REQUIREMENT that they distroy their database if necessary and enact measures to protect it from abuse. Berzerk.

1 0

Re: Voice/Fax Checks
by hfinney＠shell.portal.com 17 Dec '03

17 Dec '03

Robert Hettinga writes: >As someone who's been thrashing this a little bit, I've gotten stuck on >exactly how to "*undercut*" the transaction costs of existing methods. Got >any ideas? Are those transaction costs as a percentage of total cost >meaningful enough to embue digital cash with the rocket-like competitive >advantage we hope for? It's pretty clear that credit cards don't work for some of the transactions people want to do: 1) one-cent and fractional-cent charges for connecting to a useful Web page or ftp site. A useful resource like this wouldn't have to charge much on a per-user basis to fund the equipment and people. 2) Transactions with individuals or small companies who are not VISA clients. It's not that easy for a mail-order shoestring startup to get the ability to accept VISA cards. Because of the danger of fraud, the credit card companies like to see a storefront and/or some previous history. Someone who writes a nifty PGP shell and wants to sell it for $10 per will have this problem. 3) People who don't like giving out their credit card numbers to an unknown email address. This is the flip side of the above. The danger of fraud is always present, and the more people I've given my card number to, the more chance that I'll get burned. Of course most states have protection laws in place, but it's still going to be a major hassle. Now, 2 and 3 can probably be addressed by electronic checks, and I think the secure Mosaic announcement included that possibility. I suspect that echecks are a considerably stronger competitor to ecash than today's credit-card infrastructure. For one thing, an echeck can be sent in the clear, while ecash has to be sent encrypted; an eavesdropper can spend ecash but not an echeck. Example 1, the fractional-cent transaction, will be tough to address by any technology IMO. Even with ecash, there are a lot of questions. Is it on- line or off-line? Does the server actually try to validate each half-cent or does it just trust people? If the latter, how much fraud is likely, and how would we track down and penalize the half-cent counterfeiters? Solving these problems is going to add overhead which will make it hard to deal with such small sums efficiently. How many cash businesses sell low-value items for pennies today? Not many. >Light dawns on Marblehead. (Massachusetts joke). Isn't the point of >digital cash that you *can* send it through unsecure mail and buy things? No, I don't think you can. Ecash can generally be cashed by the bearer so it has to be sent through secure mail. That is why I was saying that echecks might be better for those purposes. I don't understand the Telescript agent world well enough to judge whether it would drive a market for ecash. I have the impression that at least with the initial implementations the agents will not be on the Internet as we know it but rather on a separate AT&T network of special servers. So they may not have much impact for a while on the "net" as we know it. Hal

3 3

Re: comeon**
by lefty＠apple.com 17 Dec '03

17 Dec '03

> Lefty says I don't have the wit to speak for him. It would take no >wit - just a person with the tastes of a bottom feeder. Lefty, aren't >you the little shit who brutally flamed Nobody for using a pseudonym? Er, no, I'm not, which renders the rest of your posting even more content-free than it would otherwise have been, if such a thing were possible. >Lefty? is that your first or last name? Do you have a last name? Are >you sure? Say, hero, what is your true name? I'd tell you, but you haven't demonstrated a need to know. Or a capacity to understand. > Lefty, do I have you all wrong, or are you one of those guys who >hang back from a fight till you see your opponent is outnumbered? Given that you're your own worst enemy, being outnumbered would seem to be a chronic situation for you. You don't qualify as an "opponent", Mr. Jeffers. Or even an hors d'oeuvre. -- Lefty (lefty(a)apple.com) C:.M:.C:., D:.O:.D:.

1 0

Re: "Key Escrow" --- the very idea
by solman＠MIT.EDU 17 Dec '03

17 Dec '03

> Eight pieces seems too few to me. It's too easy for gov't agencies to "lean > on" eight individuals or organizations (someone else suggested "watchdog" > groups as fragment holding agencies, but that doesn't seem very good. Groups > can change over time, respond to pressure. Putting a lot of fragments in a > few hands seems fairly fundamentally flawed). I'd rather see thousands. The point here is that if the evil government wants to go busting in on your conversations without a warrant, it can't. Even if they cheated and looked in the escrow for the names of you secret holders, they'd have to show them a warrant. The government couldn't try pressuring that many people before one of them blabbed and that would lose those folks doing the pressuring their jobs and quite probably result in prison time and political ramifications. I still don't like the idea of escrows because it assumes that I have something to hide, but if you have to do an escrow, I thing eight people is fine. JWS

2 1

Re: Double DES calculations
by Hal 17 Dec '03

17 Dec '03

I missed the start of this double-des thread due to system problems and being gone, and I've never been able to pick up the main point since. It sounds like some kind of meet-in-the-middle attack is being discussed. It is true that with current technology MITM generally seems more costly in terms of space than time. However, I have seen references to techniques which shift this tradeoff some, costing more time and less space. Un- fortunately, I can't remember where I saw them! I'll give you one similar example, though. I think this is the technique used in Pollard "rho" factoring. You have an iterated series, x=f(x), and you want to know if it has any cycles, any values which are eventually repeated. At first glance you might think that to look for a cycle of length N you would have to store N values of the series and check each value for a match, taking order of N in time and space. The Pollard tech- nique instead runs two copies of the iteration at once, one twice as fast as the other: x=f(x) and y=f(f(y)). Each time you just compare x and y for a match. This takes about twice as long but uses no memory. The moral is, be cautious about feeling safe against MITM attacks purely because of memory limitations. If you don't have protection on the time costs as well there may be a tradeoff which can kill you. Hal

1 0

Re: Gore's "new and improved" key escrow proposal
by gtoal＠an-teallach.com 17 Dec '03

17 Dec '03

: From: Eli Brandt <ebrandt(a)muddcs.cs.hmc.edu> : Tim May said: : > Others here will have a clearer idea than I have, but I don't think a : > "software standard" is what is now being planned. Software-only : > solution cannot possibly have the security that's needed [...] : My reading of the BSA blurb was that software key escrow really is : being planned: "software implementable [and] based on a non-classified : encryption formula". Yes, this sounds pretty silly. I don't see how : you could possibly prevent a rogue phone from interoperating with a : fascistic one. ... except by legislation. Make no bones about it, this is the start of the final attack to outlaw non-escrowed encryption. G

1 0

Re: "Key Escrow" --- the very idea
by solman＠MIT.EDU 17 Dec '03

17 Dec '03

> > ...The cost of conducting these negotiations in the > > physical world is enourmous. > > > > The cost of conducting these negotiations in the real world is negligible... > > Is "the real world" a typo? I suspect you mean something like "in > cyberspace". :) It is most certainly a typo. > I'm not familiar with the line of reasoning you're referring to here. I > suspect it's a large topic. Does it rest on the assumption that cyberspace > and the physical world are largely disjoint? > I think they're not. Activities in cyberspace often > "are about" or "have influence on" the real world. Sometimes vice versa. > Doesn't this mean laws can't be divided into those about the physical world > vs. those about cyberspace, but must in fact be about both? You are entirely missing my point. The superior efficiency of cyberspace, its low transaction costs, have created possibilities there that can not exist in the physical realm. In both realms people have different wants and desires. Because they interact with each other, the actions of one individual can have an impact upon the actions of other individuals. Without any form of social agreement, there would be no security, no certainty about anything. This can be highly inefficient. There is no point in starting a big project today, if there is a high probability that somebody will kill you tomorrow. There is no point in doing something that other people would find useful if there is no mechanism for you to exchange it with them for services that you find useful. To circumvent this inefficiency, individuals enter into contracts with one another. A and B might agree not to kill each other. In exchange for giving up this element of their freedom, they get security. Security has value. It enables them to undertake long term projects that might otherwise not have been possible. But a contract like this is not useful without some mechanism of enforcement, so A and B have to agree to pay for some sort of policeman. This policeman would receive compensation for enforcing the contract between A and B. HERE IS THE DIFFERENCE BETWEEN THE PHYSICAL WORLD AND CYBERSPACE. In the physical realm, in order for a contract like this to work, large numbers of people have to be bound to it. This is true for the following reasons: A) The number of people bound under the contract must be large enough to make it unlikely that the policeman can control them or break his contract. B) Negotiating a contract like this takes alot of time. The compensation for the policeman has to be determined. The mechanisms for disciplining murderers (and determining guilt) have to be determined. The mechanisms for enforcing the contract between the police and the people have to be determined. The mechanisms for determining how much each person will be required to pay have to be determined. This is an extraordinarily inefficient procedure in the physical world. To deal with this inefficiency we have developed laws. Laws specify that all sentient individuals within a given area have to agree to a specific contract. There is no contract negotiation, there is just a contract that automatically applies and because a substantial majority of the individuals within a given locality respect the laws, this system works. In fact legal systems are highly inefficient artifacts in the physical world. In cyberspace, the two motivations for extending contracts to all participants (i.e. having uniform laws) are no longer present. Point A is no longer true. You don't need to have an enforcing policeman whose power is balanced by the large number of people he protects, cryptography can take the policeman's place as the contract enforcer. Point B is no longer valid because the entire negotiation process can be automated by computer with negligible transaction costs. Without any motivation for laws, individuals can get their security through personal contracts with other individuals. If you don't want to worry about fraud, subscribe to a fraud protection agency that you like. Any individual agent that wants to make you an offer will have to first be approved by the fraud protection agency. In exchange for this approval, the agent will pay the protection agency money and then pass along that cost to consumers that required the approval. Alternatively, the agent can agree not to violate the rules of the FPA, and give the FPA a deposit. If the agent screws up, victims of fraud could collect recompense via the method of adjudication specified by the FPA. If the agent isn't willing to agree to the fules of the FPA, then the user and that agent just won't do business. There can be thousands upon thousands of FPAs in a scheme like this and individual FPAs can offer all sorts of protection plans. In the real world a system like this could never be implemented, the cost of administering it would be too great. How could stores know what standard of honesty was required for each individual customer if different customers lived under different laws? The store has to be notified of the specific rules governing a specific customer. The store has to pay a tiny amount to the FPA. It has to be verified that this amount is collected. The store must then alter its sales strategy for the customer. This would clearly no work in the real world. If I wanted to buy a watermelon slush outside of the Kendal T (something I am about to do) it would cost me about $200 in transaction costs beyond the $1 for the slush. This is where the difference lies. It simply becomes feasible to have individually tailored social contracts once you enter cyberspace. Individuals with incompatible social contracts simply can't communicate with one another. You get absolute freedom AND absolute security. JWS

1 0