layer hidden off the screen
UCSB English Department HomeUCSB English Department HomeUCSB English Department Home
Alan Liu Hypertext Fiction Tracker
Conceptual Demo of XML Application

Note: Proper viewing of the XML documents on this page require a browser able to render XML and XSLT stylesheets. Currently, these requirements are met by Internet Explorer 6.0 with its 4.0 version XML driver. (This page last revised 4/4/02 )

Preface (go straight to demo)

Tracker is a purely speculative, proof-of-concept demo of an XML-based "critical apparatus" for hypertext fiction. It is designed to facilitate the pedagogy and scholarly discussion of hypertext literature by providing a standard means of citing, annotating, and linking to elements in a work (no matter the work's original platform, media, or structure).

Back to lecture table of contents Next lecture section

Previous lecture section Next lecture section

Intended Uses

Authors and publishers of hypertext fiction interested in placing a work in the classroom would include an XML document describing the work and its structure in a standardized way (validated against the Tracker schema). (Alternatively, scholars and instructors would create such documents and make them available to others.) A variety of prefabricated or custom-made XSLT stylesheets (for displaying the XML document on the Web) would then allow users to cite, link to, filter, and sort works in a standard way.

In addition, XML will allow information about a hypertext work to be read automatically by application programs of various sorts. This raises the intriguing possibility of programs that provide a "meta-view" of hypertext fiction—e.g., a "fiction-browser" that creates on-the-fly views of such topics as "self," "love," "memory," etc. based on assemblages or sequences of elements from different online hypertext fictions.

XML is also increasingly convergent with structured databases (esp. relational databases). This potentially expands its use to the full range of archival and interactive information retrieval characteristic of databases. (XML can now be moved automatically in and out of structured databases—e.g., through export/import features in database programs, or XML-editing programs that can extract information from databases. XML-Query language also raises the possibility of robust querying akin to SQL for databases.)

Previous lecture section Next lecture section

Previous lecture section Next lecture section


As a proof-of-concept instrument, Tracker is very elementary and incomplete. The creation of an XSL schema fully adequate to a wide range of contemporary hypertext fiction would require the deliberation of a consortium of hypertext authors and scholars.

Currently, Tracker consists of the following components:

  1. XML document (in multiple, particular instances) containing descriptive information and links to Edward Falco's 1999 online work, "Self-Portrait of Child With Father" [link to original work]. Only a small portion of the work is now represented:

                           (Note: Netscape 4+ does not read XML)

  2. XSL schema against which the XML document can be validated to ensure standardization and completeness of data entry: hypertext_fiction.xsd (validation is performed by a script). Alternatively, a DTD could be created for the document (as in SGML documents).

  3. XSLT (Extensible Stylesheets Language Transformations) stylesheet that transforms the XML document into HTML for display on the Web: Hypertext_Fiction.xsl (and variants; see above). Alternative methods for displaying XML as HTML include cascading style sheets, data binding, and DOM scripts.

When a Web browser opens the XML document (e.g., falco_instance.xml), the appropriate XLST stylesheet is also downloaded to govern the display of the document.

Previous lecture section Next lecture section

Previous lecture section



Previous lecture section