Position Statement from DAISY Consortium, a W3C Member Submitter: George Kerscher, Secretary General, DAISY Consortium. Position Summary Accessibility for persons who are blind or have other disabilities must be woven into the fabric of all digital publishing areas. Background The Digital Accessible Information System (DAISY) Consortium http://www.daisy.org was established to identify and develop standards to make publishing accessible. DAISY Consortium Members (libraries serving persons who are blind and disabled) have had to re-publish printed works in the DAISY format to make them accessible to persons who are blind or have other print disabilities; print books are fundamentally inaccessible. However, the libraries serving the blind and print disabled have only been able to convert a small percentage of published materials. Once digital publishing sector started to develop standards, the DAISY Consortium participated in the use case identification, requirements gathering, specification development, and development of guidelines, samples, as well as the implementation of open source software for digital publishing. EPUB 3 has been developed with requirements for persons with disabilities woven into every area of the specification. Unlike other specification development, persons with disabilities are first class citizens in the International Digital Publishing Forum (IDPF) http://www.idpf.org and EPUB 3 domains. Once the IDPF approved EPUB 3 in October 2011, the DAISY Consortium endorsed EPUB 3 for the distribution of fully accessible digital publications, and the transition is underway for the DAISY Consortium Members to support and distribute EPUB 3 materials. Moreover, the potential to have all published works commercially available at the same time, and the same price, to persons with disabilities must be our goal. Handshake between Content and Reading Systems On one hand we have digital content - the EPUB 3 publication that can be fully accessible. Guidelines and a checklist are available to help content creators. On the other hand, there are the reading systems that present the EPUB 3 publication. The reading system must be fully accessible with eyes, ears and fingers, i.e. visual adaptations such as font and background, audio narration through TTS or perhaps human narration. A built-in feature to use a refreshable braille screen for content consumption by the braille reading community must also be present. In addition, all controls and access to the store or library collection must be accessible. Both the EPUB 3 and the reading system work together to provide the fully accessible reading experience. Areas of Work While much is in place right now with EPUB 3, we face interesting challenges moving forward: • Graphical content may require more than alt text. • Complex info graphics, images, etc. convey more information than can be communicated with the short alt text. A mechanism for linking to more detail such as what has been developed by the DIAGRAM Center http://www.diagramcenter.org needs to be developed. In addition, the user interface interactions must be developed. • Adaptive data visualization provides more opportunities. Traditionally, an accessible version of a pie chart would be converted to a table, probably where the pie chart started out in the first place. Adaptive data visualization also helps people who can see the graphical representation as it makes complex data more understandable. We have seen demonstrations of this by Ed Summers of SAS who showed a complex scatter graph and used regions to present the patterns that the graphical representation made clear. The regions helped the blind person and the sighted person as well. This domain of data visualization and how to make it more accessible to everybody is an area of high interest. • MathML interfaces are in their infancy. MathML holds great promise for persons with disabilities, but development of user interfaces for exploring and manipulating the MathML are immature and need further development. • Interactive content will gain in popularity in digital publishing, and this needs to have accessibility features from the very beginning. Besides further development of and proliferation of support for ARIA, open collections of accessible interactive widgets can help in making sure the interactive materials are accessible. Promotion of these open collections is another area of importance. • Reading system and app proliferation is an issue to consider. It is difficult to learn how to effectively and efficiently use an application. I believe most blind and disabled users would prefer to learn how to use a really good reading system well, and use it for most of their reading. If digital publishing evolves into individual applications for each publication, then it will be difficult for many people. We believe that separation of concerns between document content and its consumption remains a fundamental principle. Would you like to use a different word processor each time you needed to create a document?