JATS, the Journal Article Tag Suite, defines a vocabulary of XML elements and attributes used to describe the content and metadata of journal articles. As described in the previous post, we have mapped the metadata elements of the JATS Journal Publishing Tag Set to RDF, so that publishers’ XML article metadata encoded using JATS might become part of the web of linked data. Our JATS2RDF mapping document is available here in PDF format. We also created an XSLT to automate the creation of RDF metadata from documents marked up in XML using the NISO-JATS Journal Publishing Tag Library v1.0, enabling this information to be published to the Semantic Web as linked open data in a manner that is unambiguous and universally understood.
Most XML markup applied to journal articles is created by specialist companies employed by publishers for this purpose. However, to facilitate the creation of JATS-compliant metadata by others, Tanya Gray and I have separately created a JATS Metadata Input Form, by adapting the metadata input system that she had previously created to permit entry of MIIDI metadata (Minimal Information to record an Infectious Disease Investigation), as we also repurposed to create the DataCite metadata input form described in a recent post.
The JATS Metadata Input Form is freely available on the Web at http://www.miidi.org/jats. Backed by an XML model that is interpreted by XForms and Orbeon Forms to create the displayed Web form dynamically, this form contains an input tab for the five principal metadata elements that we mapped, <article> <article-meta> <journal-meta> <contrib> and <ref-list>, with an input field in each tab for the various relevant JATS metadata elements and attributes. Where appropriate, each element is accompanied by a drop-down menu that permits the user to choose one from the list of suggested input values given in the JATS specification, as shown for the element <article-type> in the following screen shot of the <article> tab:
The entered JATS metadata are saved as an XML file on the user’s local hard drive, with a name and directory location of the user’s choosing. Optionally, the metadata can also be saved in other formats including HTML, PDF and Kipling XML (a subset of the NLM Journal Publishing DTD version 3.0 that is the input format for Annotum, a publishing system based on the WordPress blogging platform). Additionally, the metadata can also be converted to RDF using the XSLT transformation described in the previous blog post.
We welcome feedback about the usefulness and functionality of this service.
David Shotton (firstname.lastname@example.org)
Tanya Gray (email@example.com)