For COVID-19 updates and guidance, visit For research-related updates, visit

A Brief Introduction to Digital Object Identifiers (DOIs)

By Ronald C. Jantz, Digital Library Architect, Rutgers University Libraries

Historically, when information such as a journal article, data set, or media file was stored in a web site or online repository, there has been concern that the link or bookmark could fail over time as the web server or site was updated.  The development of standards for creating persistent URL links to digital content has addressed many of these concerns. 

Initially released in 2000, the DOI system has become an important tool for persistently identifying an intellectual property object on a digital network.  The DOI uses a name prefix obtained from a Registration Agency (RA) and a suffix (determined by the publishing organization).  The International DOI Federation (IDF) recommends that DOIs be assigned to scholarly resources and that a single resource should not have more than one DOI.  Although the traditional use of DOIs has been for scholarly resources (primarily journal articles), there is precedent for using DOIs for other types of resources.

Getting a DOI

An individual or researcher cannot create a DOI. The DOI Federation has defined Registration Agencies (RAs) that assign DOI address ranges to digital publishers and repositories worldwide.  Those publishers and repositories then assign individual digital identifiers to objects.  For example, the citation of a recent manuscript deposited with the Scholarly Open Access at Rutgers (SOAR) repository can be shown as:

Radick, Caryn (2016). Rutgers' 1870 Centennial Celebration and Other Charter-related Puzzles. The Journal of the Rutgers University Libraries, 68(1)
Retrieved from

The citation includes the DOI identifier. The DOI in this example is 10.78282/T38G8NSM.  When combined with the DOI URL it becomes a clickable web address that takes the user to a landing page that describes and gives access to the object.

Versioning and DOIs

In science research, scholars are creating many difference types of related resources – raw data; derivatives from the raw data such as databases, spreadsheets, and diagrams; lab notes; published articles; and annotations of articles.  While a single DOI digital identifier is associated with a unique object, a single research project could be documented and findable via multiple DOI addresses, including DOIs assigned to the original published journal article(s), any Open Access (possibly preprint) versions of those articles, as well as relevant data sets and other materials.  In addition, many of these resources may undergo change resulting in new versions.  For proper scholarly use, each resource must have appropriate descriptive metadata and provenance.  These issues are more prominent with science researchers because it is increasingly difficult to package the results of their research in a single object such as a journal article or a book.  The task at hand therefore is to support the science research process and to preserve the integrity of the digital object so that it can reused at a later point in time.  Researchers, publishers, and repository managers are continuing to develop protocols for addressing the challenges of versioning information to ensure that content is not lost as new forms of information are introduced.