Skip to content

Conclusion and Recommendations

Jump to bottom
Charles LaPierre edited this page Oct 26, 2022 · 3 revisions

Introduction

In May 2022 board meeting, the board assigned an action item: "The working group on digital braille will do initial research and produce a paper explaining technical differences and non-technical pros and cons of the identified approaches in order to inform decision making concerning DAISY ongoing involvement."

To achieve this, we formed an ebraille working group consisting of braille experts from DAISY members and community. The experts collectively developed a problem statement to make sure that all of us have a common understanding for the problems that we want to solve with the new braille format. Then the experts gathered the use cases which the new braille format needs to address. The use cases led to a set of technical requirements. Then the working group evaluated which of the existing standards can meet these technical requirements and which standards provide a shortest and most efficient path for meeting these requirements. The next section explains the conclusion and recommendations in detail.

The technical analysis document is a living document and will be updated along with the further refinement of the use cases. But it is sufficiently mature at this time for providing the recommendations for the direction of the project.

Conclusion and recommendations

Important remarks:

When we say that a standard is not able to meet a requirement, it means the inability of the standard in its current state, it does not mean that the standard cannot be extended to meet the requirement.

The braille formats like BRF, BRA, BRL etc. are represented by the term text based braille formats in this document.

Standalone content document

As explained in the technical analysis document, we evaluated PEF, text based braille formats, DAISY XML (DTBook.xml) and XHTML/HTML standards for this requirement.

The analysis of the technical requirements derived from the use cases makes it clear that there are existing standards which can be leveraged for this purpose and we would not need to develop new standards from scratch. By extending a well-established standard we will be able to cut down considerable amount of efforts involved in developing a new standard, developing production and reading tools and the adoption of the standard.

The next question is which of the existing standards provide us the shortest and the most efficient path to the success. It involves evaluating which of the existing standards meet most of the technical requirements and making a judgement if there are some major issues which can block extending any of these standards.

The analysis makes it clear that the existing braille formats like PEF, text based braille formats etc. do not support most of the technical requirements for the content. It is possible to extend the existing braille formats for this purpose, but it would require significant amount of work. On the other hand, the digital text publishing formats like HTML, DAISY XML (DTBook.xml) are much closer to meeting all the technical requirements for the content. Therefore, extending digital publishing formats like HTML, DAISY XML etc. is the recommended approach.

The next question is, which one of the digital publishing formats should be our first preference?

HTML/XHTML and CSS are very widely used on the web as well as in publications. It will help in quick adoption and integration with the main stream technologies, many engineers around the world know these specifications well and software and libraries for processing HTML and CSS are widely available. On the other hand, extending these specifications according to our needs is challenging. HTML is maintained by WHATWG, which is driven by the browsers and adding the features which do not have the web as the primary use case is an uphill battle. CSS is maintained by W3C and we would face similar challenges in extending it. We do have a work around for this problem, DAISY Consortium can create a customized version of HTML or CSS and add the new features in a way which allows web browsers to ignore these additional features and render it as a web page, while the specialized reading systems can make the use of these additional features.

DAISY XML (DTBook.xml) is well known in the DAISY community, it is owned by DAISY Consortium so extending it is comparatively straight forward, it is not a mainstream specification but the engineers in DAISY community know it quite well and many people in DAISY community have created software and libraries for processing it. On the other hand, its scope is limited to DAISY Community and it would be challenging if we plan to integrate it with the main stream technologies some time in future.

Therefore, we recommend leveraging HTML/XHTML and CSS for this work and keep DAISY XML (DTBook.xml) as a back-up. In case we hit a blockage in extending HTML/XHTML and CSS specifications according to our needs, then we can fall back on extending DAISY XML.

Set of content documents weaved into a publication

For weaving HTML/XHTML content documents together into a publications, we can leverage any of the existing mechanism’s like EPUB 3, Web publications, HTML index file etc.

Remark: It is important to clarify that EPUB 3 has a different purpose than XHTML. XHTML contains the content of a publication whereas EPUB 3 is a container which packages XHTML documents, provides reading order, metadata etc.

Metadata

Metadata for the braille publication is another essential component of our work. We recommend leveraging the Dublin Core for publication metadata and schema.org for accessibility metadata. We will need to work on the braille specific metadata, some of which is for the use of the readers, and some of it is for giving instructions or information to the reading systems.

Packaging specification

We need to provide an option to package the file set of the braille publication. We can leverage any of the following existing specifications:

Braille graphics specifications:

we need to figure out if the requirements of multiple line braille displays can be fulfilled by some of the existing image formats or if we need to start work on a new format for displaying graphics through braille dots. This is a vast topic and needs more research. This may be delegated to a dedicated task force which works in close collaboration with the reading systems or reading device developers and provide the recommendations.

Clone this wiki locally