Appendix B: On The Horizon

The date: 30 June 1996.

So what's the future of HTML look like today?

Let's take a moment to survey the land ahead of us, so to speak, and get a sense of where things will be going in the next year or so. (Note that I am skipping subjects such as VRML, as these are very complicated subjects whose future is clouded at best. VRML, for example, may totally revolutionize the Web in the next year-- but we've been hearing that for the last two years, so I remain a bit skeptical.)

Updating HTML-- Finally!

The current version of HTML in widespread use is HTML 2.0, which has been the standard for nearly an eternity (at least two years). There were a number of new tags which appeared during that time. These tags were created by Netscape, Microsoft, and others for their own browsers, but the core of HTML is the universally supported 2.0 tags. For a while there was an initiative to create HTML 3.0, which would be more advanced, offer more tags, and so on.

The 3.0 effort was disbanded last year in favor of a number of smaller initiatives-- one for tables a separate one for math markup, yet a third process for new style tags, and so on. It was thought that by making each new feature its own initiative, simple features could be added quickly, while more complicated things, such as mathematical markup, could proceed at a different pace.

However, this approach was found to be unsatisfactory, mostly because all of the initiatives-- even the simple ones-- moved at a pace which can be described as somewhere between dead slow and dead stop. So there is once again a unified initiative to update HTML; the version under discussion right now is HTML 3.2.

In more detail, HTML 3.2 includes a number of "Netscape tags" and "Microsoft tags;" that is, tags which were first supported by either Netscape Navigator or Microsoft Internet Explorer because the browser authors had invented these tags. This is apparently being done in an attempt to account for the multitude of pages already in use which employ these proprietary tags. (The wisdom of this approach may be debatable, but it's where the community is headed.)

So what can we expect from HTML 3.2? To name some of the new features: backgrounds and text colors, as supported by Netscape; greater alignment control over elements such as paragraphs; the ability to create document divisions with separate rules for each division; the CENTER tag; more control over the appearance of horizontal rules; tables; strikethrough text; big and small text; superscripting and subscripting; more image attributes; APPLET and OBJECT containers; and client-side image mapping.

Whew! And that isn't everything the new specification includes-- it's just a look at most of the highlights. What does it all mean? For authors who have been writing Netscape-specific pages, and users of Netscape Navigator, the answer is, "Nothing much will change." Tables, backgrounds, and so forth, will all look very familiar to those knowledgeable about the Netscape tags, and a few Explorer tags made their way in as well.

It should be mentioned that one major Netscape tag-set which is not part of the HTML 3.2 specification is frames. Complicated to implement and often confusing to the user, frames are likely to retain their Netscape-specific status for quite some time.

The Web Gets a Sense of Style

Related to all of this, but not exactly part of HTML, is an effort to define a way of setting up style sheets for Web documents. With style sheets, an author is able to define very specific rules and assert a great deal of control over the fine points of document display, as opposed to the fairly coarse level of control offered by current Web browsers.

The specification for Cascading Style Sheets, Level 1 (CSS1) is out, and it offers the author approximately 35 areas of control over document appearance, in an almost unlimited variety of combinations. Some of the things authors can specify is font colors and styles (uppercase, title case, small-caps), background colors and patterns, borders, spacing, indentation, and font families, to name just a few.

An interesting aspect of CSS1 is that it defines the author's style sheet as carrying more weight than user's defined style. In other words, according to the CSS1 specification, if the user has defined a standard style sheet for his browser, and he loads a page which has an associated style sheet, the loaded style sheet will override the user's defined styles. Whether or not browser authors follow this paradigm remains to be seen, but it could make for some fascinating debates in the Internet community.

What's It All Mean?

In the short term, the new HTML tags will have little apparent effect, since most of them are already in widespread use and supported by the three most popular browsers. Where the real change will be realized is in CSS1, which will allow for the creation of truly hideous-looking pages... but it will also allow authors to very elegantly present useful information. Look for and praise the latter, and ignore the former. The Web's future depends as much on the praise of users as it does on the effort of authors.

Of course, this whole Appendix will be out of date within the next nine to eighteen months. That's as it should be-- the rapid development and maturation of the Web is one of its key strengths.

Next -- Appendix C: References
Previous -- Appendix A: Tag Summary
Table of Contents
Glossary
Index