You're sitting at your Macintosh, trying to remember where you saw that picture of the new Porsche you liked so much. You log on to your Internet Provider, a network of Sun SPARCs running an Apache server on top of Solaris. Firing up your Web Browser, you switch to one of your favorite Web sites, a complex search engine driven by a vast array of HP Alphas running Linux. You type in your search string and the list comes up. Hmmm. You still can't remember where you saw the picture. You click on what looks like a likely prospect and....there it is....the object of your desire, sent courtesy of a single PC in Germany running IIS server on top of Windows NT. You save the .jpg file to a drive on your Macintosh, making a mental note to turn your prize picture into a screen saver in the morning.

While transferring files from one world to another over the internet may seem almost childishly simple to the casual observer, the reality of the situation is quite different from the appearance. The World Wide Web is host to a dizzying array of computers, operating systems, server protocols, data types, and compression schemes, all carefully routed through God-knows-how-many switching networks and thousands of miles of wire and fiber optic cable. The immediate popularity of the Web is almost entirely attributable to the skill with which the software engineers have hidden the complexity of the system from the users. Making the contents of your Web site available to your audience and maintaining the illusion of simplicity is going to require some thought and advance planning on your part.

    In this section we're going to raise some of the basic technical questions that you're going to have to answer before your Web site can go on the air. Like most multimedia projects, designing a Web site involves wading through a seemingly endless sea of compromises.....style vs. compatibility, image quality vs. download time, glitz vs.content, and of course, budget vs. everything. We will eventually develop a compatibility profile which will guide us through the development phase, but before we do that, we have to answer a larger question...

How Much Compatibility Is Enough?
    The answer to the question of where to draw the line on compatibility issues is one which will affect almost everything else in the development process. The capabilities of Web browsers are arranged in a kind of loose hierarchy starting with the modern browsers, which have powerful suites of handling routines for everything from secure e-mail to virtual reality modeling, and going back to the early browsers, many of which are text only. While the new technologies available on the Web allow you to do things that you couldn't even dream of a few years ago, they generally require powerful computers and fast connections to work properly, requirements which are sure to leave a portion of your audience in the cold. While we would expect that implementing the latest technologies could be a little tricky, the truth is that even plain old text can get us into trouble. To find out why, let's take a look at the Web's most basic language...

HTML
    HyperText Markup Language is an offshoot of another language called Standard Generalized Markup Language, and is the basic language of the World Wide Web. HTML is based on ASCII Text, which means that you can create HTML with almost any word processor. One of the principle reasons that HTML is so powerful is that it is Device Independent. This means that the HTML files that you create on your PC for your Web site will probably run just fine on a Mac across town or a UNIX machine on the other side or the world. The HTML instructions, called TAGS, are Interpreted by your browser and translated into machine code written specifically for your operating system.

    Standardized General Markup Language, HTML's parent language, is still used extensively, mostly by large institutions. In general, it is a robust and more complex language than HTML. Both languages allow the creator of the document to communicate his intentions as to how the information should be handled, while leaving a large number of specific formatting decisions such as the choice of font, font size, background color, etc., to be determined by the person viewing the document. For instance, in HTML, if I wanted to Emphasize a portion of text, I would use two tags. First, I would use the Emphasis tag <EM> to tell the browser to begin emphasizing text until I tell it to stop. When I had emphasized as much text as I wanted, I would tell the browser to stop emphasizing with the </EM> tag. These two tags together form a CONTAINER, telling the browser to perform the same operation on all the text contained between the beginning and the ending instructions. Because HTML operates using universally agreed-upon standards, my intention to emphasize a particular section of text is communicated regardless of the font, font size, or background color that you choose.

    This method of dealing with text files has certain advantages, not the least of which is that it virtually eliminates the digital Tower of Babel that we've become so accustomed to......

....I send you a letter written in Microsoft Word XP, but your copy of Microsoft Word 6 won't open it. You send me a file written in Word Perfect. My copy of Appleworks will open it, but it's filled with so many extra characters that it looks like someone fell asleep on the keyboard. The formatting is ruined....

Because large institutions are almost universally forced to deal with such formatting conundrums on an immense scale, SGML seemed like an almost perfect solution. The problem with SGML is that it's a relatively complex language. As more and more text files were transmitted digitally, the need for a simplified universal method of transmitting and interpreting text files became more apparent, and so HTML was born. Early versions of HTML followed the philosophy of the "SGML School", leaving many of the formatting decisions to the end user. The subsequent explosion of the Web brought a large number of content providers from the world of desktop publishing, where they had become accustomed to having an enormous amount of control over formatting issues. As HTML has evolved, the clamor for more control by members of this "Author" school has been answered by the development of a fairly large number of additional TAGS. These new instruction sets provide site designers with both the multimedia capabilities they need to handle a broad spectrum of file types and the formatting tools they need to generate Web pages with a more magazine-like appearance. These improvements are not free, however. As browsers become more and more sophisticated, more platform related conflicts arise, creating a force which acts in direct opposition to the notion of Universal Access, the very idea which made the Web popular in the first place.

    Riding herd over this entrepreneurial circus is the W3 Consortium. The Consortium is led by teams of scientists at the Massachusetts Institute of Technology, the French National Institute for Research in Computer Science and Control (INRIA), and KEIO University in Japan. These three organizations make up the core of the governing body, which is charged with defining standards for HTML and hopefully preventing a runaway proliferation of "features" from deteriorating into intergalactic anarchy. The remainder of the W3 Consortium is primarily made up of scientists and theoreticians from over 150 organizations spread throughout the world. The standards approved by the consortium define the feature set which must be supported and adhered to before a browser or a Web page can be said to be compliant.

This brings us to another subject which has been a thorn in the side of a large portion of the governing body for many years now, namely......

The Browser Wars
    Encouraged by its early lead, Netscape Communications led the charge to develop new uses for the World Wide Web. Based on the original Mosaic browser developed by a team of students at the University of Illinois' National Center for Supercomputing Applications, and driven by a legion of creative insomniacs armed with mountains of feedback from the users, Netscape pretty much redefined the way the Web was used around the world. Netscape's Communicator, which was designed to incorporate a "plug-in" architecture based on open standards, laid the groundwork for seamless integration of HTML, secure commerce, and mail with a myriad of transfer protocols, peripheral computer languages, and multimedia file types. Netscape flung the doors open to outside developers and challenged them to develop new applications for the Web.

    Microsoft, somewhat belatedly recognizing the notion of open standards as a threat to their market position, introduced a whole other set of slick Web-based features and development tools that are only available to viewers using Microsoft browsers and operating systems.

    Because adherence to the official HTML standard by definition imparts a certain parity among software developers, Netscape and Microsoft's development and support of non-standard features was seen by some to be an attempt at an end-run around the governing body in order to gain market share. This tail-wagging-the dog scenario does seem perilous. If everyone did this there would be chaos. And surely almost anyone can recognize the danger of allowing any single company with a clear corporate agenda to dominate an entire world market.

Since many of Netscape and Internet Explorer's most popular features are accessed through the use of non-standard tags, this dual path development war has caused somewhat of a schism among developers and viewers .

The W3, in an effort to steer the Web world back toward digital convergence, abandoned the development of simple HTML after version 4.01, and replaced it with a new, reformulated and expanded language called XHTML. The X in the name means that the language is extensible. It opens the door to a rich set of custom features that are once more based on open standards rather than proprietary code, and should go a long way toward helping developers leap the operating system and application protocal barriers to inter-enterprise connectivity.

As it turns out, Microsoft, through the sheer weight of its monopoly in operating systems, has now pretty much marginalized Netscape and many of the other competitors to Internet Explorer. Even though it is pretty much wagging the dog at this point, it has, to its credit, done a really nice job of making Internet Explorer comply with most W3C standards. However, IE's nagging security problems have opened the door to a newcomer, Firefox, which is being developed by Mozilla.org as an open source project fueled by tank cars of caffeine and an international army of gifted volunteer programmers. For now, it is a safer alternative to IE, is powered by the exceptional Gecko rendering engine which is fully compliant with W3C standards, and has an open, plug-in architecture which will encourage entrepreneurs to develop new uses for the browser.... at least until Microsoft kills it with Longhorn. (For the record, the terms of a recent Justice Department consent decree now require Microsoft to say "pretty please with jam on top" before devouring a competitor.)

That being said........

Here Is Your Dilemma:

You're developing a Web site. You really love some of the propriatary features available to Internet Explorer users, but you're afraid of leaving millions of your potential viewers out in the cold. What DO you do?

    You'll find some of our suggestions in the Recommendations section. In order to help answer the question for yourself, try this.....download as many of the different browsers as you can find on the net. Most are free or at least offer a free trial period. Pick a short list of your favorite sites, then visit each one in turn using different browsers. Be sure to take notes. I guarantee that the experience will be very educational.