A standards-compliant archive format for research.

Introducing DAR

Research is currently created using print-oriented file formats such as DOCX or LaTeX. However, a modern published article requires the availability of structured metadata such as authors, affiliations, references and journal information. Publishers typically make use of the JATS-XML standard to capture that information, resulting in the costly and slow conversion process to turn a DOCX submission into structured XML.

If researchers and publishers all used a shared file format, documents could be created and shared directly without friction. Furthermore submission, peer-review, editing and author proofing workflows could operate on the same source document, streamlining the entire publication process.

A Document ARchive (.dar) is essentially a zip archive, containing the manuscript as a standard JATS XML file, as well as additional assets such as images and data files. It is built around the following design goals:

Structured article content is stored as JATS, the de facto standard for the archiving and interchange of scientific open-access content with XML.
DAR is designed for machine readability. This makes it easy for a range of tools to be developed (e.g. PDF conversion, MS Word import/export etc.)
All kinds of assets, such as images, data files, additional documents are directly included in a DAR file.
It is possible to include custom content through Texture plugins. For instance, Stencila uses an extension to offer code cells.