The components of a database may be any informational material, but the significant distinction, regarding copyright law is between those components that are protected by copyright and those components that aren’t protected by copyright. In addition, in order to become an eligible creation, copyright law requires that creation to be ‘original’. Hence, the fundamental classification or typology of databases requires that the constituent parts of a given database are examined and a decision is made as to whether the assemblage of its elements is original within the meaning of copyright law. Thus, a database in question should be examined in two dimensions: content and structure.

The first distinction concerns the database content. The database elements can be one of two categories, namely works or data. A work is a copyright subject matter, while data is considered to be in the public domain. These two categories are not distinguished by their intrinsic nature, but by the applicable copyright law.

The second dimension relates to the database structure. The database structure can be either original or non-original, depending in the test of ‘originality’ in any particular jurisdiction. In this discussion, the above-mentioned distinction is named as the distinction between ‘set’ and ‘stack’.

A ‘set’ means that the collection of elements constitutes a distinct unit. In this sense, the database is ‘original’ in the meaning of that given term with respect to the applicable copyright law. On the other hand, a ‘stack’ means that the elements put in the database are not assembled in an ‘original’ fashion, when ‘original’ is defined within the meaning of the applicable copyright system.

Four categories of databases are available regarding this two-dimensional classification, which relates to the elements of the content and to the structure of the collection. The four categories may warrant distinct treatment in copyright law and are discussed below.

• Work-Set: original collection of works

In this case, copyright subsists in the components and another distinct copyright in the collection as a whole. Certainly, incorporation of works into that type of database requires authorisation by the respective owners.

• Data-Set: original collection of data

As defined above, the components are not subject to copyright protection. However, the database, as an original work, is copyright protected. The issue here is to which extent extraction of the non-protected materials is allowed.

• Data-Stack: non-original collection of data

Arguably, databases of this type cannot be protected by copyright. The fundamental requirement of ‘originality’ is missing. The absence of protection to databases of this category may require alternative modes of protection.

• Work-Stack: non-original collection of works