Supporting multiple media

November 20, 1997

Seybold Report on Internet Publishing
Vol 2, No 3

Until now, the Web has been stuck with a delivery format that was essentially second-class when it came to data representation. Its advantages—timeliness, interactivity, and low cost of delivery—are offset by crudeness of formatting and a set of tags whose semantics were intended for journal articles, not the richness of the full spectrum of publishing.

With XML we at last have the potential for text and data to be portable across print, CD-ROM and online media. We had that with SGML, of course, but the only way to deliver it over the Web was to "dumb it down" to HTML or to view the source data with an SGML viewer. Neither of those are optimal solutions; the former lacks the semantic richness of the source file; the latter often shows more than desired (assuming you’d like to show customers a subset of the full tagging of your source files). With XML, the SGML publisher has the same workflow as with HTML, but with a much more robust target, one that could make use of a publisher’s unique tags.

In a conference session discussing cross-media publishing, Paul Beyer of Banta and Paul Trevithick of Archetype described how metadata serves as the basis for organizing and managing document components, and how XML data, with design models, templates and style sheets, can feed a new generation of tools for creating both online and printed documents—without the big tradeoffs in quality that one might expect.