Web Applications Working Group,
Greetings. At the upcoming Digital Publishing Workshop, topics will include
"widgets definitions, standardization"
(http://www.w3.org/2012/08/electronic-books/topics.html) and, depending upon
the workshop, "widgets definitions, standardization" could pertain to web
applications, web components, or both, and web components could become an HTML
5.1 topic.
As digital book authoring teams are envisioned as making use of advancing
authoring software with visual design premises, digital books and textbooks are
anticipated to contain nested visual design components in layouts. Such visual
design components can be implemented as web components.
With regard to hypertext-based documents, books and textbooks containing nested
web components, topics of interest include bookmarking to and hyperlinking to
specific configurations of books, chapters, sections, and pages in multimedia
hypertext-based documents, digital books and textbooks.
With regard to nested web components and multimedia elements, there is a tree
of stateful objects concept and bookmarks could be implemented as XML files,
for example in a BookmarkML format. Such XML data could be obtained from and
loaded into digital book readers' JavaScript API's.
BookmarkML could include, for example, XPath paths to nested stateful web
components and multimedia elements as well as initialization data for each web
component and multimedia element.
<bookmark>
<data path="...xpath1...">init_string_1</data>
<data path="...xpath2...">init_string_2</data>
</bookmark>
or
<bookmark>
<data path="...xpath1...">
<param name="x1">value_1</param>
<param name="x2">value_2</param>
</data>
<data path="...xpath2...">
<param name="x3">value_3</param>
<param name="x4">value_4</param>
</data>
</bookmark>
With a BookmarkML format, hypertext containing a hyperlink to a page in a
digital book could resemble:
"...as you can see in Book, on <a href='book.page123.bookmark'
type='application/bookmark+xml'>page 123</a>..."
With regard to a URI syntax, based upon the tree-based nature of the data, it
is possible that a predicate calculus syntax could be utilized in URI
fragments, as per: #P1(X, P2(Y, Z)) .
With BookmarkML, URI syntax, or both, pertinent topics include the interfaces
of some stateful hypertext elements, <object> and multimedia elements, and
heuristics could be described to obtain data from hypertext elements.
Kind regards,
Adam Sobieski
