counter-example, the blood vessel pattern on the surface of a kidney, while perhaps unique to each person, would probably require invasive techniques to measure. In addition to being non-invasive, any measurement equipment should be relatively inexpensive and highly reliable. It should be possible to certify a level of trustworthiness of the equipment.

When a person enters an identification system, a copy of their differential identity marker is captured and stored in the identity registry for the system. This process is termed enrollment. At the time a person enrolls, the marker can be compared to all existing markers in the registry to confirm that the person is only entered into the system once. This process is termed identification. The process of capturing the marker from the person and comparing that to a copy of the marker that was stored away when the person enrolled in the identification system in question is termed either verification or authentication. The marker should be of such a form that a highly reliable comparison of captured versus stored markers can be made. The probability should be low that markers that should match don’t, and the probability should be high that markers that should not match indeed do not match. The comparison process should be relatively quick, which means not really more than a second or so. If it takes two hours to perform the comparison process then the authentication operation will be useless for most interactions.

We can also suggest a couple of characteristics that would allow us to keep differential identity teased apart from privacy. First, the marker should convey no information about the individual other than the differential identity of that person. Moreover, the marker should offer minimal or non-existent forensic evidence. In other words, it is desirable that the marker not be amenable to unintended indication of physical presence, which is an example of forensic evidence, except through an intentionally activated authentication protocol.

Following this rather lengthy prologue, we will now cut to the chase and assert that at the present time, the most reliable differential identity marker mechanism that possesses these characteristics is one or more of the biometric traits of the human body. Principal among these are fingerprints, handprints, facial characteristics, retinal patterns, iris patterns and DNA. Further, it should be noted that we’re suggesting that an actual biometric characteristic forms the differential identity marker, not some card or other token that encompasses the biometric characteristic.

The last two marker requirements that we noted illustrate our concern for the demarcation between establishing pure differential identity and using differential identity as an index for other information that we often want to relate to identity. As an example, if we want to adhere to our set of stated marker requirements, we will be more inclined to use a fingerprint or an iris scan versus using a DNA pattern because DNA conveys much more health related information that an individual may well want to hold private. Moreover, because we can derive DNA from blood or other bodily fluids, the presence of DNA is an ambiguous indicator of physical presence. That is, since blood can be taken from an individual and then placed at some location where that individual has not actually been, it is a potentially erroneous forensic indicator. We will be more inclined to use an iris scan versus a fingerprint because the iris scan offers significantly less forensic value than does a fingerprint. When we handle things, we often leave behind our fingerprints. When we look at things, we rarely leave behind our iris patterns. However, iris patterns contain more health related information than fingerprints.

Book available at Midori Press (regular)
Book available at Midori Press (signed)
Book available at Amazon (regular)