A Quaint Discrepancy between My Dated Meaning of ‘Index’, and the Current Meaning

I actually became acquainted with computers, doing a data-entry job, on a long-forgotten database management system, on a long-forgotten PC. According to what I remember, a database consisted of a main table of records, each of which had a fixed set of fields, plus any number of Index Tables. Each record in the main table had a record number, which was only meaningful to the one DBMS. Each index table consisted of the values of one particular field of the main table’s records, sorted, and each associated with one main table record number, in which the value of the indexed field occurred.

This meant that if a look-up needed to be performed according to which up to 3 fields had a defined value, notionally, the first of those 3 fields could be looked up in its index table, while the additional 2 fields would have been considered to occur in random order, and their legitimacy determined through comparisons, that all resulted from the look-up of the first field, in its index table.

Today, an index is said to refer to something else. If the blocks in a database have a block-number, then the main table is thought of as a kind of ‘array of blocks’. The index, or block-number, is the index of this array.

Dirk