| Previous | Table of Contents | Next |
The increasing usage of electronic delivery for documents has led to increasing sophistication in their features and functionality. A major trend in this direction is a movement to tailor the document structure, content, and formatting according to the specific profile of the reader.
To support this, the SGML document may be tailored through a variety of techniques, including entity usage within an electronic document’s stylesheet (as well as within the SGML document instance itself).
Dynamic documents can also be tailored to the user through the use of different stylesheets or the support of system variables in stylesheets.
• See “Using and Creating Output Specifications in Panorama Pro,” p. 455 for an example of the use of different stylesheets on the same document.
As you may have surmised, there have traditionally been problems in defining standards for output specifications. Although SGML has offered great flexibility in the use (and reuse) of structured information, the mechanisms for defining output formatting have tended toward the vendor-specific side.
The result has been that you could develop an SGML system for the reuse of transportable information, only to be stuck with spending a great deal of time and effort in the preparation of vendor-specific output specifications for your documents.
Fortunately, the problem of vendor-specific output specifications is being addressed through the development of output specification standards. Beginning with the development of the FOSI and continuing with the advent of the DSSSL, the ability to support industry standard output specifications is starting to take shape.
The following sections examine several standards for common output specifications.
The Format Output Specification Instance (or FOSI) was an early attempt at achieving a standard for an output specification. Developed early in the output standards process, it was an attempt to facilitate output formatting that was not vendor-specific.
As an early attempt at an output specification standard, it is limited in its functional capabilities. Perhaps its most important specific contribution is its ability to handle element formatting according to the element’s specific context in the document structure hierarchy. As you saw earlier in the chapter, this capability facilitates flexibility in handling element formatting.
For example, context sensitivity allows you to format titles occurring within chapters differently from those occurring within appendixes.
Limitations in the FOSI led to an initiative for an approach offering greater flexibility in an output standard. A committee was formed to develop an ISO standard (ISO 10179) for a Document Style Semantics and Specification Language (DSSSL).
The desire to offer much greater flexibility and capability in the DSSSL (when compared with the FOSI) has resulted in a standard of considerable complexity. Thus its development has spanned a number of years since the establishment of the DSSSL committee in 1988.
DSSSL specification consists of two major parts:
The first part (for structural transformations) will serve to facilitate additional processing steps that traditionally have not been easy to accomplish with prior output specification standards. Prior to DSSSL, these steps may have required external processing by dedicated programs. These may include:
Formatting capabilities supported by DSSSL include the recognition of an object’s context (as with the FOSI) but are much more comprehensive. Processing features supported include the capability to recognize specific content within the SGML document instance and structuring output formatting based upon that content.
Because of the complexities involved in the development of the DSSSL standard, an interim version (or subset) has been developed. Known as DSSSL Lite, it focuses on the structural formatting portion of DSSSL. By providing a subset of the structural formatting portion of DSSSL, DSSSL Lite is intended to be easy to include in current generation SGML software products. It supports both electronic and hardcopy formatting, although the composition capabilities of the latter are limited.
As the standard functionality defined in the DSSSL (and DSSSL Lite) standards are adopted by SGML vendors, the process of developing output specifications will become both easier and more standardized across SGML processing environments.
In this look at output specifications, you have examined their relationship to the document content model. You also looked at issues involved in providing output formatting for SGML document instances, including those that apply to hardcopy and electronic versions of a document. The chapter closed with a look at the current state and future direction of output specification standards.
For more information, refer to the following:
| Previous | Table of Contents | Next |