VI · 39

OpenURL & DOI

DOIs identify resources; OpenURLs deliver users to a context-appropriate copy of them. The two technologies are complementary, not competing.

The DOI (Digital Object Identifier) and OpenURL solve different problems. A DOI gives a resource a stable identifier that can be resolved to a default URL. An OpenURL carries a citation — possibly including a DOI — together with the context of the request, so that an institutional resolver can choose an appropriate destination. Both are NISO/ISO standards and both are in widespread use in scholarly publishing.

Carrying DOIs in OpenURLs

A DOI is carried in the rft_id parameter of an OpenURL using the info:doi/ URI form:

rft_id=info:doi/10.1103/PhysRev.47.777

The info:doi/ namespace is registered (RFC 4452) and is the standard way to express a DOI inside the Z39.88-2004 framework. Resolvers that recognise the namespace can use the DOI as a primary lookup key; those that do not can fall back to descriptive metadata if it is also present.

The DOI proxy accepts OpenURL syntax

The DOI proxy at doi.org — operated by the International DOI Foundation — accepts OpenURL queries in addition to direct DOI lookups. An OpenURL pointing at doi.org will be parsed; if a DOI is present in rft_id, the request is redirected to the DOI's registered URL. If only descriptive metadata is present, the proxy can in some cases match it to a known DOI through CrossRef metadata.

https://doi.org/openurl?
rft_id=info:doi/10.1103/PhysRev.47.777

How they complement

  • The DOI provides identity: this resource, however it is hosted, in whatever form.
  • The OpenURL provides context: this user, at this institution, with these subscription rights.
  • An OpenURL including a DOI is the best form for a link in a citation: it carries the identity for the resolver that recognises the DOI, and the metadata for the one that does not.

Sources