In the spirit of practical science, let's test various DatabaseDefinition against these candidates for "database" or DBMS:

'''Candidate Definition Abbreviations:'''
* AMP - TopMind's "Attribute manager/processor common idioms" definition
* RF - DatabaseIsRepresenterOfFacts 

'''Array'''
* Integer Indexed
** AMP: No - covers insufficient idioms
* Associative
** AMP: No - covers insufficient idioms

'''Persistent Array'''
* Integer Indexed
** AMP: No - covers insufficient idioms
* Associative
** AMP: No - covers insufficient idioms

'''File System'''
* AMP: No - Does not primarily deal with "attributes", but rather documents and binary blobs.

'''Windows Registry'''
* AMP: Borderline - covers insufficient common idioms

'''LDAP'''
* AMP: Yes

'''Text Indexing Engine (such as Google)'''
* AMP: No - Does not primarily deal with "attributes", but rather documents.

'''Expert System'''
* AMP: No - Expressions are not "attributes". Otherwise, they'd call them "expert bases".

'''The Cyc AI engine'''
* AMP: No - Expressions are not "attributes"

------
See Also: DatabaseDefinition, DatabaseIsRepresenterOfFacts