The Digital Publishing Interest Group had its first face-to-face meeting during the W3C TPAC week in Shenzhen, China. The meeting’s main goal was to give more direction to the various task forces that the Interest Group had started to define in earlier weeks, by specifying their scope and main focus. Not all task forces were covered; indeed, the two days also included meetings with other experts from other W3C groups, so the final scoping of the task forces will have to be done in subsequent teleconferences.
The issues around pagination (based on the very fist draft document) were, obviously, the most complex. Producing a document covering all aspects for all kinds of publishing would be a huge task, going beyond what the IG could reasonably do. After much discussion, the scope, for the first version of the document, was restricted along two axes: publishing genres and writing systems. For the former, some of the possible areas were, for the time being, were set aside; these are journals and magazines, poetry, children’s picture books, comics and manga. Indeed, these genres require very special considerations (e.g., possibly fixed layout), and additional efforts and resources will be needed to cover those. As far as writing systems are concerned, the group had to take into account that W3C already had published documents on Japanese Layout, and similar documents may become available in future for Chinese, Korean, or Indic writing systems. As a consequence, the current document will concentrate on Latin based pagination, including the various local variations for different languages or cultures.
Beyond the current pagination concerns (i.e., headers, footers, page breaks, etc.), it was also recognized that typography issues, again concentrating on Latin languages, should be considered to be very much in scope along the same lines as pagination. Whether this will be a separate document or part of the pagination document is still to be decided.
Although the pagination work primarily results in issues around CSS (possible missing features, setting priorities, etc), which was also the subject of a joint meeting of this group with the CSS Working Group, it was also recognized that pagination raises a number of problems in terms of the content model, in the DOM, as well as available events (e.g., event should be raised when user turns a page). These notions are necessary for reading systems, and it is not clear at this moment whether all the necessary features are covered by the current set of events defined for HTML and/or whether DOM extensions would be necessary. A separate document will have to be published to look into this, which may result in some further joint work with the HTML Working Group in the future.
A very different problem area the group looked at is what is currently known as “Behavioral Adaption”, exemplified by some use cases on the IG Wiki. The solution of those problems require some sort of an additional markup identifying, e.g., the publisher’s semantics for specific elements (chapter title, index, etc.). There are different approaches: one is to use more powerful metadata syntaxes like microdata or RDFa Lite to annotate the content; the other is to use e-book specific attributes as extensions to the core HTML5 set. After discussions on the pros and cons of these two alternatives, the IG decided in favor of the attribute approach. This will be considered in more details in the months to come. The current EPUB specification already introduces an EPUB namespace, yielding
epub:type attributes; however, that approach may lead to issues in the future in view of the evolution of HTML5. The direction that will be explored further is the attributes of
epub-XXXX format, i.e., without the usage of the XML namespace syntax. It was recognized that a document specifying these attributes, as well as possible values, should be produced (probably by IDPF) to get this accepted as a bona fide HTML5 extension.
The issue of security was also addressed. After quite some discussion it was decided that this large area of concern should be made more specific to decide what is, and what is not in scope for the Interest Group. The issue of DRM on books naturally came up; indeed, it would be, in theory, possible for the IG to collect use cases for various forms of DRM. However, the feeling was that the IG would never get to a consensus on such use cases, due to the different appreciations of the underlying business models. As a consequence, the IG decided that DRM is out of scope for this IG. There are, however, other security as well as privacy issues that are relevant for digital publishing: e.g., what happens if a malicious URI is added to the spine of an electronic book, what happens to the private data a reading system may collect on the user’s behaviour, etc. These issues are very much in scope, and the decision of the IG is to explore those areas further.
There were other discussions areas, sometimes with guests coming from different groups within W3C, e.g., on accessibility or testing. The more detailed minutes, both for the first day and the second, are available on line.
It was a good meeting, which also gave the possibility for many to meet personally for the first time! Additionally, members of the Digital Publishing Interest Group attended other working group meetings throughout the week which, hopefully, was useful for everyone involved.