S.L.:

...

also, my question about whether one loses the downloaded Chaumian bucks if one's hard drive crashes was answered in the affirmative by Levy.

But as I said, if the drive is backed up you don't lose it. It's my understanding this is the case with the current Digicash/Twain bucks. But not necessarily with all implementations.

you apparently didn't notice but there were some other posts on this list about writing down the random seed that a person used to generate the cash, and then being able to restore the cash somehow based on knowing the random seed, however privacy is sacrified in this case apparently. this with the Chaumian implementation, from what I understood. if this is true, I wanted to point out in my post that this is a property that real cash does not have. you cannot ever say to your bank from which you withdrew cash, "oops, I lost a piece, could you replace it? also, if someone spends the piece that I lost, tell them they can't?" how would the bank handle a situation in which the consumer says, "oops, my hard drive was compromised. all of the following cash was NOT spent by me." notice this is precisely what happens with stolen credit cards today all the time. the bank, if it was real cash, would have to say, "sorry, you're out of luck. it's your responsibility to keep your cash secret". or, the bank might say, "oh, well, that cash was not spent yet. we can issue a replacement and bar the cash from being spent". this idea of "replacing" cash is of course unique to the electronic realm. if the bank does it, the consumer would be required to notify the bank before the cash is spent, again another unusual property unique to e-cash. I would suggest to all companies developing e-cash that they print up a carefully designed FAQ in which their cash properties are compared directly with parallel properties of credit cards such as like the item above, because the consumer is of course very familiar with credit cards. (this will be very crucial for longterm customer education and acceptance imho. otherwise the customer may make assumptions like, "since the cash comes on a card it behaves like a credit card", etc.) I would like to see someone elucidate what happens with lost Chaumian cash in a detailed post (the oneliners I saw so far don't address the complexity and seriousness of this issue. this is a really significant issue with digital cash IMHO that will become very obvious once people who were used to the "mushiness" of credit card transactions switch to cash). there seem to be many caveats with this "restoration". one immediate problem I can imagine: the recent netscape bug showed very well that a short random number generation process can be hacked. so in other words, if the random numbers are short enough to write down, it makes me wonder about the security of the cash generated from this seed.

This "lose your money" with digital cash really does seem to get to people. I always point out that when you take $ out of an ATM machine, you don't expect to get it back if you lose it. For those forms of e-money that are irretrevable when lost users are warned not to download huge amounts.

yes, it is clearly an emotional issue. the problem with a lot of technology is that people are very emotional and irrational, and hyperactively so when it comes to spending money. all marketing campaigns are designed to capitalize on this. various competitors can be tainted merely by cheap psychological ploys that have no basis in fact. (the back and forth between AT&T, MCI and Sprint and other competitors is an example of the endless mudslinging, marginally based on reality, that can ensue.) as I wrote in my earlier post, the key is to let people know of the weaknesses in the product ahead of time so they are prepared. and the analogies that you offer are precisely the kind the company can give to the customer to make them feel more comfortable that "this stuff is more convenient than real cash and not really any more dangerous". it's critical that they do so ahead of time however, before the customer gets zapped by his own naivete.

Question: will people's worries about losing their e-money lead them to accept a higher degree of tracibility as a tradeoff?

that's exactly the issue I tried to raise in my post, unfortunately perhaps in not the crisp way you do here. (of course, you're the one that makes a living off your prose, heh.) one can see advertising as a kind of yin/yang game. you have to imagine that if your competitor says, "gosh, you could lose cash with so-and-so's system", you can always reply "yes, but you retain privacy-- can you do that with so-and-so's system"? digital cash is not a panacea. it is in fact a design compromise that simply applies more importance to various aspects of transaction capabilities: ease of use, privacy, etc. checks and cash are at two ends of the economic transaction spectrum. in contrast to other cypherpunks I don't really see either as intrinsically superior to the other. many businesses may *never* end up using digital cash in certain transactions (i.e. with other companies, not individual buyers) for traceability reasons. it seems to me there are many agreements in which hiding the identity of one or both parties is simply intrinsically unacceptable to both, and no amount of handwaving smoke-and-mirrors appeals to magic "reputation agencies" will completely alleviate this. so my bottom line is that I think digital cash is going to become a very, very important part of the future economic ecology, but it will coexist with other methods. it may very well become the dominant method, but I suspect there will be significant areas of commerce that actively resist digital cash technology out of the preferences of all parties. something the hardcore cypherpunks in general fail to consider, that was brought out in a great article in the WSJ about the counterfeit $100 superbills circulating: the cash aspects of the global economy are negligible in comparison to "identified" transactions such as through credit and checks. (hence the treasury's lukewarm approach to stopping counterfeiting, which scares the willies out of me personally). there is absolutely an *enormous* dedication to "identified" transactions in the world today, and cypherpunks seem to me to be somewhat misguided in the ways they exalt cash. it is true that cash is the medium of choice for the individual, but the vast majority of economic transactions are *not* made by the individual in virtually any civilized society. digital cash may change this slightly, but not significantly, I suspect. businesses exchange money as part of complex agreements and contracts in which liability is a key ingredient. liability is an exceedingly problematic issue with digital cash. if anyone is going to create a widespread cash economy based on total anonymity they have an uphill battle easily far more difficult than Chaum has encountered so far. (remember Chaum only implements "semi-anonymity", anonymity on payer side but not payee side). -- anyway, congrats on the fine article SL. you get my "honorary cypherpunk of the week" award for the article <g>

Ok. So we've got Alice and Bob trading their cash back and forth, with Mallet trying to steal their cyber-dosh, and Eve listening to their conversations- this leads me to the big question: What's the right name to use for the Banker? I've been using "Nick", after Nick Leeson; any other suggestions? Simon --- (defun modexpt (x y n) "computes (x^y) mod n" (cond ((= y 0) 1) ((= y 1) (mod x n)) ((evenp y) (mod (expt (modexpt x (/ y 2) n) 2) n)) (t (mod (* x (modexpt x (1- y) n)) n))))