HTML 4.0 Sourcebook
(Publisher: John Wiley & Sons, Inc.)
Author(s): Ian S. Graham
ISBN: 0471257249
Publication Date: 04/01/98

The Microsoft approach is similar, but different. Microsoft has implemented an extension of their TrueType font format, called OpenType. This contains compressed TrueType fonts, which can be distributed and referenced in a manner similar to TrueDoc fonts, although the font file formats and the underlying technologies are incompatible. With Microsoft’s technology, the desired, downloadable fonts must be referenced using style sheet rules of the form

@font–face {font–family: family–name;
            src: url(<http://domain.zz.com/fonts/fontfile.eot>’);
           }

where family–name is the name to use in the document to reference this font.

Once again the two approaches are incompatible. The Netscape approach has the advantage of font flexibility, since any font can be used when authoring. The disadvantage is rendering speed, since the software that draws the fonts is built into Netscape Navigator 4 and is not terribly fast. Microsoft’s approach is much faster at drawing text, as the downloaded OpenType fonts are converted into TrueType fonts, which are rendered very quickly directly by the operating system. The major limitation is that the fonts used must be TrueType or OpenType, as other formats are not supported.

Missing Features: Mathematics

HTML is currently missing a number of useful hypertext document description features. Notably absent is a way of expressing mathematical symbols and equations. The absence of mathematical expressions is particularly surprising, given the origin of HTML in the scientific community. At present, mathematical equations can only be included by using a separate application to create GIF images of mathematical expressions, which are then included within the HTML document.

Using TeX and LaTeX on the Web

In the mathematics and scientific community, the LaTeX typesetting/document formatting language is almost universal for preparing scientific and mathematical texts. For this community, Nikos Drakos of the University of Leeds has produced a sophisticated program that can convert a LaTeX document into a hypertext collection of HTML documents, automatically converting mathematical expressions into inline GIF images. This remarkably useful package is called latex2html and is available at:

cbl.leeds.ac.uk/nikos/tex2html/doc/latex2html/latex2html.html

Of course, this means you need to learn LaTeX as well. For less sophisticated needs, there is a convenient collection of GIF–format Greek letters and other mathematical symbols, which authors can include in their documents to create simple mathematical equations or scientific expressions. A crib sheet for these icons plus access and licensing information (there is a charge for commercial use—read the license agreement for more information) can be found at:

donald.phast.umass.edu/latex/tutorials/kicons.html

www.anachem.umu.se/graphics/symbols/symbols.html

donald.phast.umass.edu/kicons/license.html

TeX/LaTeX Viewer Plugin

A third alternative is to use a browser plugin that can format and display TeX/LaTeX documents. IBM has developed such a tool, named techexplorer, available for download and evaluation at:

www.ics.raleigh.ibm.com/ics/techexp.htm

This package does not support the full set of TeX/LaTeX markup commands; rather, it supports a large subset, plus special extensions for hypermedia and windowing support. Thus, LaTeX documents originally prepared for print publication often need some reworking to be viewed via techexplorer. The documentation at this URL describes the techexplorer–supported TeX/LaTeX language and some of the steps that this adaptation process may entail.

Other Mathematical Alternatives

The current preference of the HTML language design community is to leave mathematical expressions out of HTML. The goal is then to create a separate markup language that is mathematically parsable—that is, a markup language that contains sufficient understanding of the meaning of the expression that the expression can be used as input to mathematical analysis programs, such as Mathematica or Maple, or spreadsheets, such as Lotus 1–2–3 or Excel. The World Wide Web Consortium has been actively pursuing this approach and recently announced a working draft of the proposed MathML mathematical markup language. Interested readers should obtain the working draft listed in the references at the end of the chapter.

Other Missing Features

Other annoyingly missing features are page banners (nonscrolling headers or footers for documents—these can be mimicked by frame documents, but not completely reproduced), footnotes (perhaps displayable as pop–up items above the regular text), tabbing control, and the ability to mark arbitrary spots in a document (for linking or referencing purposes), or a range of text in a document (for selective retrieval of document fragments). Several elements (FN, TAB, SPOT, RANGE) in support of these functions were proposed in HTML 3, but have subsequently been dropped from the W3C proposals for the next generation of HTML. With luck they will be reintroduced in the future, as such functionality is often needed in a general–purpose markup language such as HTML. For additional reading on these topics, you are referred to the expired HTML 3 draft document and to the Next Generation HTML Starting Points listed in the references.

XML—A Successor to HTML?

One problem with HTML is its inflexibility—every change to the language must be approved by a large, unwieldy committee, such that even the slightest extension to the standard can take years to approve and implement.

Recently, however, a new language was developed that foreshadows the possibility of customized markup, tailored to each application. This new language is known as XML, for the eXtensible Markup Language. Unlike HTML, XML does not define a set of tags or elements. Instead, it defines a scheme whereby authors can create their own elements and attributes, allowing them to define document structure as they wish, without the limitations implicit in the element structure of HTML. This document can then be distributed and processed by any software equipped with an XML parser (the software that structurally processes the markup). To make this latter step easy, XML was designed to be simple so that such parsers are easy to write.

Indeed, simplicity was one of the main goals of XML. In some ways, one can think of XML as a “simplified” version of SGML, retaining all the best features of SGML while discarding those features that are rarely used and overly complex to implement.

XML is already widely used on the Web—for example, the Channel Definition Format (CDF), used by Microsoft’s “push” channel to distribute HTML content to subscribers, is written in XML. In the future, when XML documents can be distributed along with a style sheet to specify how the XML elements should be displayed, one can expect to see XML being distributed in much the same way as HTML.

Keeping Up–to–Date

Many of the details about upcoming features of HTML are found at the World Wide Web Consortium Web site, listed at the beginning of the references section. If you’re burning to know what’s up and coming, or you’re just being thorough, this is the place to look.

References

Next–Generation HTML Starting Points

Netscape Frame Documentation

home.netscape.com/assist/net_sites/frames.html