Skip to main content
ExLibris
  • Subscribe by RSS
  • Ex Libris Knowledge Center

    How does incorrect metadata break OpenURL linking?

    Question: How does incorrect metadata break OpenURL linking?

    Answer:

    OpenURL linking relies on matching metadata to link from the referring citation to the provider's platform. Incorrect metadata from a referring source, in any augmentation source, or at a target platform can all break OpenURL linking as it will often fail to match with correct metadata from other sources.

     

    What is OpenURL?

    OpenURL is a provider-neutral syntax used by link resolvers to connect users from a citation (referring source) to the full text of an article, book, or book chapter (target).
    For more information on OpenURL, please see the NISO standard.

     

    How does OpenURL linking work?

    Referring sources (such as Primo or Summon) are repositories of citation metadata. To connect a user to full-text, the referring source first sends an OpenURL link containing the relevant citation metadata to the link resolver.

    Link resolvers (such as SFX, Alma link resolver, or 360 Link) take the incoming OpenURL and use the embedded citation metadata to:
    •    find a title match in the link resolver’s knowledge base
    •    find a portfolio or holding’s match in the library’s subscriptions
    •    construct an outbound link using the citation metadata to the target’s subscribed platform

    The outbound link to the target will organize the referring source’s original citation metadata using a provider-specific syntax (often called a “parser” or “linker”). The user will then arrive at the full text of the cited article, book, or book chapter.

    As an additional feature, SFX, Alma link resolver, and 360 Link will overwrite or augment the referring source’s citation metadata with metadata from CrossRef or PubMed if the OpenURL has a DOI or PMID. This helps ensure we are using the most accurate and up-to-date metadata to connect your users with full-text.

     

    How is OpenURL linking impacted by incorrect citation metadata?

    Each step of the link resolver functionality depends on citation metadata, so incorrect citation data can cause links to fail in several ways.

    An incorrect title or ISSN/ISBN in the referring source citation can cause errors with the knowledgebase matching logic, and the link resolver may match to the wrong item or fail to find a match. When this happens you may get a “No Results Found” page from the link resolver. (This can also happen when the title-level data in the knowledgebase is incomplete.)

    If the article title, author name, start page or other metadata in the citation from the referring source does not match the same field at the target, you may get a “No Results Found” page at the target platform. This can happen either if the referring source metadata is correct and the target metadata is incorrect, or vice versa.

    Finally, if an augmentation source, such as PubMed or CrossRef, has incorrect citation data in their systems, it can cause a mismatch between the output URL metadata and the target metadata that was not caused by the referring source metadata. 

     

    What can we do about broken OpenURL links?

    When there is an OpenURL failure we will investigate and troubleshoot it with you. If the root cause of the failure turns out to be incorrect metadata in our systems, or if there is a workaround we can implement using the metadata in either our discovery or link resolver systems, we will work with you to update the metadata or linking. If the root cause of the failure is due to incorrect metadata in an external system, and there is no workaround available to us, we will explain that to you. At that point you may wish to contact the third-party vendor to request they make a correction to their metadata.

    If there are recurring patterns with incorrect metadata from specific providers, we will work with those providers to update their discovery metadata. However, as this process relies on working with external parties, we cannot guarantee specific timelines for resolution.

    For more information about causes of OpenURL linking failures, please see: What Are the Common Causes of Full Text Linking Problems, and How Can Linking Be Improved?


    • Article last edited: 1-Nov-2019