A taxonomy (Taxonomy (Dictionary Entry)) is defined by bailey (Bailey, 1994) as a classification of empirical (observable) entities. In that it is distinct, if not the opposite of a typology that is a classification of concepts (c.f.: ).

To enable the construction of robust identity taxonomies, it is thus necessary to establish an inventory of empirical (observable) entities, to arbitrage if these entities are of the class identity or not and to describe these entities.

In order to progressively establish this inventory, a table structure is provided below. The approach shall then consist in progressively referencing items found in the field and/or the literature.

There is a necessary catch-22 here: to designate the empirical entity, we need to use a term and that term itself will be a type (that is a concept). We thus need to find a balance and use classes of empirical entities that are as atomic and as unambiguous as possible, while not being unnecessarily fine-grained. For instance, a local user of an OS family may be an interesting object to consider because it has unique properties. But it may not be necessary to specifically consider the local users of every version of that OS family as distinct empirical entities. Conversely, the generic class local user may be too vague in view of the diversity of shapes and forms it may take in certain OS families.

The more exhaustive this inventory will be, the more exhaustive will be the taxonomies based upon it. Thus, in order to privilege exhaustivity over consistency, this inventory is not composed of mutually exclusive classes but overlaps should be documented to facilitate the analysis of the inventory.

In view of the number of documented empirical entities, the inventory could not possibly be exhaustive but at least it should become representative of this wide variety.

To privilege concision and readability, definitions are provided as link to dictionary entries.