The metric I propose is to measure the number of “private data records” that exist for a particular individual, i.e. that contain information that the individual would consider to be “private”. Additionally, we should measure how many groups/people have access to these data records, and how many of these groups/people share or access the information without the individual’s consent. Other important factors to measure are: the average lifetime of a particular record, and the amount of time required to access a particular record.
For example, in 1860, before the advent of computer databases, governments and organizations kept data records about individuals, usually in the form of paper files. Such records would include criminal records, birth and death registries, and marriage records. However, the number of such records would be limited by practical considerations such as the space requirements of keeping voluminous paper stacks. In addition, much of this information, such as birth records and marriage records, would not even be considered “private” by the individuals concerned, since a birth or marriage was usually publicly celebrated in the local community. Also, because of lack of copying technologies, the records could not be easily shared or distributed, thus limiting how many people/groups had access to these data records. Because of a limited amount of space to store paper, old records would necessarily be discarded after a certain period, thus limiting the average lifetime of a particular record. The amount of time required to access a particular record would be high, as manual indexing systems would need to