Authoring accessible ‘Tagged PDF’ documents using LATEX

Several ISO standards have emerged for what should be contained in PDF documents, to support appli- cations such as ‘archivability’ (PDF/A) and ‘accessibility’ (PDF/UA). These involve the concept of ‘tagging’, both of content and structure, so that smart reader/browser-like software can adjust the view pre- sented to a human reader, perhaps afflicted with some physical disability. In this talk we will look at a range of documents which are fully conformant with these modern standards, mostly containing at least some mathematical content, created directly in LATEX. The examples are available on the author’s website, http://web.science.mq.edu.au/~ross/TaggedPDF.

The desirability of producing documents this way will be discussed, along with aspects of how much extra work is required of authors. Also on the above website, and published elsewhere in this issue (pp. 131–135), is a ‘five-year plan’ on how to modify the production of LATEX-based scientific publications to adopt such methods. This will involve cooperation between academic publishers and a TUG working group.

Editor’s note: Since the talk worked mostly from examples, showing non-printing aspects of what can be stored in, and extracted from PDF files, the printed description is not entirely sufficient; see the video at youtube.com/watch?v=mPBtkCsChJw.