2.1 Standardization.  Below is an illustration of how the query set is reduced through standardization. The data comes from a research problem in Norway. The assignment of a standard to the various spellings of a given name reduces multiple queries on that name to one. Similarly multiple queries on the patronymic, which was used historically as a surname, may be reduced to one. So if both the given name and the patronymic are needed to identify a person, and the person making the query doesn’t know how the name will be found in the record, there are many possible queries. Some of the possible ways to record a certain man’s name appear in the following chart. Here there are n possibilities for the given name: GN₁, …, GN_i, …, GN_n. Similarly there are m possibilities for the surname: SN₁, …, SN_j, …, SN_m. This means there are n * m total possibilities to form a query. Standardization reduces this to one.

It is easy to see that the problem is much larger even than this example, when all the variant spellings are taken into consideration. The surname “Taylor” in the British Isles has more than 150 spellings alone. The place called “Wuerttemberg” has been found referred to in any of over 2500 different ways of being spelled. The given name “Elizabeth” in all its variations numbers over 4,000. Such large n values imply a very cumbersome query management.

In actual practice the form of the authoritative standard is irrelevant. Hence, if the query comes in for “John Tailor” of “Wurtemberg,” the standard forms may be “Johannes Schneider” and “Wuerttemberg.” But, the standard might also be some set of codes, such as, “00003487” for the given name, “00236349” for the surname and “007830” for the place.