Blog
Knol

is quite clear; the population is to be enumerated, not estimated. So, it generally takes about a year to perform the count. It costs billions of dollars and the count requires the services of about a million people for some portion of that year. Alas, the count is wrong every time! The basic reason it is wrong is because the process used for the count doesn’t do the two things that our pre-kindergarten curriculum says we have to do to count a bunch of things. Specifically, the process doesn’t include a good way to establish a differential identity marker that we can associate with each person so that we can in the future determine whether they’ve been counted or not. In addition, the counting process used is comparable to our example method for counting apples, except that we’re hard pressed to get the entire population of the country to stand in their appropriate boxes.

With a large population that is continuously in a state of flux, we suggest that the best count that one can do is to asymptotically approach the correct answer; that is, to determine the actual number of people in the sets to be counted. Therefore, a case can be made that if accuracy is the goal, then with the same amount of work, we would be better off maintaining a continuous count for which we can constantly improve the accuracy and precision, rather than restarting the count from scratch every ten years to reacquire information we should already have. However, to so accurately count the population of the United States, we would need a permanent personal differential identity system that could maintain a continuous count of the population. One might argue that this is the correct mandate for the Bureau of the Census. Perhaps that’s a topic best left to a different discussion. However, such a system would allow us, relative to any single person, to accurately determine whether they’ve been counted or not and it is certainly germane to our current discussion to consider the methodology of such a system. It is, in fact, central to the concept of extending the mechanisms of human social orders into cyberspace.

While the United States Census presents a moderately difficult case, virtually any situation in which we want to be able to establish the differential identity of individuals presents the same set of issues. Moreover, we have yet to consider the problem of linking experiential identity to differential identity in a trusted manner. So, to address these issues in more detail, let’s consider the issues associated with differential identity markers.

When faced with counting a large, constantly changing population, care must be given to the manner in which we make a correspondence between the individual person and a set of numbers that have counting as part of their being. In particular, if we’re going to actually trust the final tally that we arrive at, we need to make sure this correspondence isn’t easily counterfeited or misinterpreted. Historically, the approach used in establishing such a marker has been to issue a credential to the person that asserts the differential identity of that person relative to other persons within the set from which the credential derives. Such credentials are typically issued within some group, so each credential is aimed at distinguishing a specific person within that group. This, of course, harkens back to the examples we noted earlier under the guise of experiential identity. In fact, a credential is actually a formal record of a discrete experience of experiential identity. A major difficulty with the attempt to use credentials for establishing differential identity is that the credential is typically only ambiguously associated with the person, perhaps through physical possession or perhaps through societal convention. We have already discussed at some length an example of a credential that is associated with an actual person through societal convention; their name. As we also noted, a name is a moderately reasonable credential to be used in the authentication of differential identity in a small group. A credential might also be associated with a person through physical possession. Thus, a person might hold a credential in the form of a token, for example a signet ring. Using a more modern example, a useful credential might actually be a hybrid of the two, such as a card with the name printed on it (let’s say that there is no picture

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