Persistent Identifiers

The data curation lifecycle approach to managing research data seeks to avoid the loss of valuable research data through ineffectual management or short term planning.  Sometimes, however, the problem of long term access to data arises not because of a lack of planning, but because the location of a particular resource changes.  This can occur because of a change in standards or software at the facility that manages the server or repository, or as a knock-on result of a parent page being either changed, relocated or taken out of use.   It can also occur as a result of ‘bit rot,’ or other technical failure of platform without strong technical backing.  Over time the risk grows that documents etc. will no longer be accessible at the location given, e.g. in a reference, due to a change of the document’s electronic location. This resembles the situation of a unique book in a library that is not on the shelf at the position indicated in the catalogue.

Persistent identifiers seek to provide some reassurance in the face of this possibility.  The increase in the authority of digital resources (by ensuring they are stable, and can be consulted as a part of investigating the provenance of a research argument), as well as their reach (by making the resources visible to search engine crawlers) enable the permanent identification and location of Internet resources, be it a single publication or a set of files.  They enable the permanent identification and location of an Internet resources, be it a single publication or a set of files.

Standards for Persistent Identifiers

There are a few different standards for persistent identifiers.   One common one is the DOI (Digital Object Identifiers), which are globally unique, alphanumeric strings assigned to one digital object. It is permanently assigned to a specific electronic resource. A DOI is attached to a digital object in a way that lasts, and is managed apart from the object’s physical location. The DOI itself never changes. This provides the advantage that a properly managed DOI will always point to the current online location of the object, and can therefore be used reliably in all references and citations. Persistent identifiers provide a lasting reference to a digital object. They can be actionable, providing a persistent link to the digital resource.  They allow datasets to be tracked and cited. By doing so, they encourage access, discovery, and potential reuse of datasets.  As such they are important for research, and are often incorporated into research infrastructure services (such as the handle PID supplied to any documented deposited in the DARIAH HAL repository.)