🚀 go-pugleaf

In article <ARI.HUTTUNEN.92Sep9013737@supergirl.hut.fi> Ari.Huttunen@hut.fi
(Ari Juhani Huttunen) writes:

>It just shows that the current idea of an 'ATM card' is no good. We need
>intelligent tokens and pretty damn fast. By 'intelligent token' I mean
>a card that can be challenged by the computer of the bank. I've seen this
>talked about in many books, so it's nothing new. What I would like to know
>is that how soon such devices will be available?

Once upon a time I worked for a bank.  I read a report estimating
losses on credit cards from fraud vs. other causes, as well as
detailing some of the security measures on today's cards.

The long and the short of it is:  Losses due to fraud are currently
an order of magnitude less than losses due to simple non-payment
of what's owed.

Until fraud becomes a more significant problem, there is little
motivation to invest resources into reducing fraud when the same
resources could be used to try and reduce non-payment of debts.


At the same unnamed bank, we were required to use smart cards for
some things, and heavily encouraged to use them in some pilot
customer-type projects.  This was two years ago, and maybe state
of the art has improved, but the smart cards we were using had a
lot of reliability problems (chip popping out of card, not working
after you sat on your wallet, not surviving exposure to water,
etc.).  Customers already get very upset when their mag-stripe
cards stop working, and replacing them with something that is much
more fragile doesn't seem like a good idea.


So smart cards are likely to be delayed until costs fall nearer to
the costs of making a mag-stripe card, and reliability increases
to a level similar to a mag-stripe card.

Or until fraud becomes a more serious problem...

- Paul Colley
University:  colley@qucis.queensu.ca
Home:        pacolley@ember.uucp   watmath!ember!pacolley  +1 613 545 3807
<Ring> [...] "Sorry, I'm all booked up." "Who was that?" "The library." - B.C.