Establishing identity in order to punish, threaten or intimidate
If some third party identity-establishment service were instead to guarantee that they would track down and prosecute the identified party in case that party committed fraud, then that kind of identity establishment would have meaning. The NRC report on authentication, "Who Goes There?", calls this Authenticating to Hold Accountable in Chapter 2.
If we lived in the old days when people stayed within a few miles of where they were born, just knowing who the fraudster was would have been good enough because he wouldn't get away without giving up his entire life. Today those conditions don't apply, so knowing precisely who the fraudster is has no meaning unless we have the ability (and money) to capture and prosecute the perpetrator.
And, of course, this all has no value if what is lost through that fraud can not be made whole. If lives or secrets are lost, no amount of prosecution after the fact would make up for the security breach.
If we lived in the old days when people stayed within a few miles of where they were born, just knowing who the fraudster was would have been good enough because he wouldn't get away without giving up his entire life. Today those conditions don't apply, so knowing precisely who the fraudster is has no meaning unless we have the ability (and money) to capture and prosecute the perpetrator.
And, of course, this all has no value if what is lost through that fraud can not be made whole. If lives or secrets are lost, no amount of prosecution after the fact would make up for the security breach.
posted by cme at 12/07/2004 01:13:00 PM
4 Comments:
This is a really important discussion.
I believe that one of the problems we're running into with identity in modern society is essentially non-technological: the identities of the real people we deal with today are often very indistinct compared to the identities our ancestors - even recent ancestors - were used to.
In small villages and even big cities a century or two ago, the average person habitually did business with people who were well known to him either from birth or from a long history of daily interaction. Most people travelled seldom and almost never moved.
Today the situation is very different. We interact with many people whom we never meet, and even more people whom we meet only a small number of times.
You & I (Carl) have met in person - what? - probably fewer than 20 times in our lives, and yet we're fairly close friends. I've lived in 9 different cities and almost 20 different houses or apartments. I've never lived ANYWHERE for more than 15 years.
Given this, I think that a solution to problems of identity has got to start with a recognition that identity is not very firmly established even in the real world at the current time.
Credit agencies, background checks, letters of reference, etc... are all essentially introduction services - they attempt to intialize a new acquaintance's view of my identity with information which is derived from a community which has a more or less firmly established view of my identity. But these views are pretty easy to game if you're willing to go to some trouble (move, grow a beard, get a sex-change operation, etc...) - and ONE discontinuity creates the possibility that I will tailor a new identity to some fraudulent purpose, patiently building up the proper dossiers until I can spring the trap (insert evil laughter here).
If you don't believe that introduction services are a good solution, I think you begin to move away from identity, and toward designing transactions which mitigate risk in ways not related to identity - for example by creating symmetrical risks for all parties to the transaction. Even this has limitations, however, as suicide bombings demonstrate...
Bob, you prompted me to count how many places I've lived:
longest in one residence: 9 years
22 residences
14 different cities
8 different states
Bob,
I believe introduction services (reputation services) are essential. My problem is that they got busy doing that work and apparently just assumed that authentication would happen automagically. A credit bureau gathers reputation of allegedly someONE - but it might be a number of people. They aggregate information and most of the time the aggregation is correct, but how do they know?
Even if the aggregation is correct, if I walk up to some new lender and claim to be Bob Blakley and present his SSN and other information, will they be convinced that I'm he and that the new account they issue me should be aggregated under his credit report? ...that he should be billed for my purchases?
This wouldn't happen if we had a way to authenticate in the physical world as securely as we do in the computer world (even though that isn't as secure as we'd like it to be).
I'm bothered that few people (if any) are giving active thought to how to authenticate better in the physical world.
...and to forestall some suggestion that we just issue a National ID card for better physical world authentication...
When I lose that card and have to go to another card issuing station to get a replacement copy, how will they authenticate me? It won't be via the card I've lost.
If we come up with some way to authenticate without the card, we could use it now - and skip issuing the card.
Post a Comment
<< Home