Abstract
This paper examines the need for stronger foundations in XML schema design, particularly when working with complex content models such as DocBook, TEI, or VRA[7]. These models are based on element and attribute relationships, requiring validation scenarios that enforce structural and semantic accuracy. While significant effort goes into developing schemas to represent these complex models, their accuracy in capturing data requirements is often assumed.
To address this, the Content Architecture team at Bloomsbury Publishing have introduced a unit test-driven methodology for validating the structural integrity of Bloomsbury XML schemas. Our initial implementation of this schema test suite has proven effective in supporting schema modifications by providing reliable feedback to verify the reliability of our schemas.
Table of Contents
We thank our fellow Bloomsbury Content Analysts and Content Architecture colleagues who contributed to the testing infrastructure and schemas that made this project possible.