Insos's Dyna Text professional publishing system

Emedia Professional, July, 1999 by Leonor Ciarlone

synopsis: Inso's DynaText Professional Publishing System publishes and delivers highly searchable, SGML/XML-based electronic books to LAN, CD-ROM, Internet, intranet, or extranet applications. This end-to-end publishing solution uses automated, repeatable, and maintainable processes to apply structure and formatting to word processing-based documents, thereby making the promise of one source to multiple outputs a reality for commercial and corporate publishers.

It used to be that publishing was relatively easy, given the fact that the primary--and often only--output deliverable was a printed document, rarely reused by either the publisher or user. At this point, we all know the Internet has changed that. And along with the change has come an increased demand for more than just a static, paper-based product.

Given the power of today's desktop authoring tools and subsequent digital delivery formats, the good news is that now everyone has the power to publish an electronic document. The bad news is that everyone does. Even worse, everyone does it differently. In many organizations, multiple word processors, various formatting criteria, manual conversion processes, and an inordinate number of humans generate LAN, CD-ROM, and Internet-based products at a cost so high it may not even be quantifiable. With this scenario, the dream of one source being channeled reliably to multiple outputs is confined to myth.

With a different scenario, one that places the burden of conversion and output on technology rather than humans, the dream becomes more a tangible reality. Add the possibility of delivering electronic documents with rich structure, multiple views, intelligent searching, and powerful linking, and the reality becomes a competitive advantage.

the reality of SGML/XML publishing

Inso's DynaText Professional Publishing System publishes and delivers highly searchable electronic books to LAN, CD-ROM, Internet, intranet, or extranet applications. This end-to-end publishing solution uses automated, repeatable, and maintainable processes to apply structure and formatting to word processing-based documents, thereby making the promise of one source to multiple outputs a reality for commercial and corporate publishers.

Built on the open, vendor-neutral, Standard Generalized Markup Language (SGML) and Extensible Markup Language (XML) standards, DynaText brings the power of these technologies to organizations whose infrastructures rely on proprietary word .processing, but whose document-based products require both depth and portability. The foundation of the product's success is based on the core SGML and XML philosophy: the separation of structure, content, and format. The reality of the product's success is that results can be achieved without agonizing conversion at the authoring and editing stages.

The system is composed of three primary applications. DynaTag structures, formats, indexes, and publishes Word, FrameMaker, Interleaf, and WordPerfect documents to compiled SGML or XML electronic books. The DynaText browser serves the book to LAN or CD-ROM clients. Its partner, DynaWeb, dynamically serves books to the Web as chunked HTML files via a standalone Web server or server module. The system uses a Unicode-based architecture to support publishing and viewing for international languages that include French, Italian, German, Spanish, and Japanese.

building a rules base with DynaTag

Even if you've never heard the "separation of structure, content, and format" mantra, you'll appreciate the ability to take a set of proprietary documents and repeatedly produce multiple outputs as a turnkey operation. DynaTag's promise of a reusable, automated publishing environment rests on its ability to convert streams of incoming word processing documents (styled content)to SGML or XML (semantic, structured content). The promise is achievable given a controlled, style-based authoring environment. However, like the old adage, "garbage in, garbage out," the effectiveness of the DynaTag output, and in turn the usefulness of the electronic books, is directly related to the consistency of its input.

Your primary goal in DynaTag is to create a "rules base" that defines the structure of the elements or "objects" in the electronic book(s). The task is as simple or complex as your viewing and searching requirements. Treated as an application development exercise, the resulting project serves as a batch-processing center for word processing input, to be maintained and updated as the nature of the content changes. DynaTag's Project Mappings report provides a foundation for the system documentation you'll need to support the application.

DynaTag offers a graphical user interface that outshines its predecessors. The three-pane view of the input objects, output objects, and content provides a clear association between incoming styles and outgoing SGML/XML elements. A "data repository" pane organizes word processing formats into logical groups, lists the details of each mapping rule, accesses the Style Editor, displays the resulting SGML tagging, and lists a log of all errors within the session.