When one is concerned about an object identification so as to make objects recoverable via their identity, a system of UniqueIdentifiers is usually devised. Whether a collectable signed and numbered by the originator, or a file name in a filing system, or an object or method in an ObjectOriented system, or a pattern. Those who are concerned about OnceAndOnlyOnce will necessarily be found utilizing methods of identification which are unique.

Other more numerous reasons for UniqueIdentifiers can be found in:
* http://www.ietf.org/rfc/rfc4122.txt
* http://www.ariadne.ac.uk/issue8/unique-identifiers/

Universal Identifiers
*http://en.wikipedia.org/wiki/Universally_Unique_Identifier
Use GloballyUniqueIdentifier''''''s (GUIDs) in Access and SQL Server
* http://www.devx.com/dbzone/Article/10167
When does a Unique Identifier not need to be used? ItDepends on Context:
* http://www.networkworld.com/newsletters/dir/2005/0530id2.html
Also, there's the 'tag' URI protocol, which I first saw proposed to be used within AtomSyntax for handling unique ids:
* http://www.faqs.org/rfcs/rfc4151.html
* http://taguri.org/
----
Related:

Contrast with GroupsAreNotUnique
----
CategoryOrganization