The Future of the Web

by Joe Gillespie — Jul 1, 1998

I suppose I'm pretty lucky.

Not so long ago, a National lottery was introduced in the UK with weekly jackpots of eight million pounds or more.

The chances of winning this jackpot are zillions to one but I thought that I would be clever and write a little computer program to improve my chances.

Being a crazy designer and virtually number-dyslexic, my thinking was more mystical than scientific. Most "normal" programmers would not even think of including the response to 'What is your favourite colour?' in an algorithm, but nevertheless, the program did help me win some small prizes. That had nothing to do with my programming prowess, but more to do with the vast number of tickets I bought to test my theories.

Predicting what is going to happen with the World Wide Web is probably just as futile. Every day, some newspaper or magazine columnist paints a vision of how the Web is going to mature into some kind of utopian, all encompassing organism that will do everything for everybody. There are even people who believe them!

Although I have done in the past, I'm not going to make any such predictions now, but just take a peek around the next bend.

The Benefits of Hindsight

I had my first encounter with a home computer in the early eighties. In those days they all had some form of BASIC programming language and it was possible for any Tom, Dick or Harriet to write programs. But BASIC was, and still is, a very slow interpreted language. With the need to produce fast, arcade-style games, programmers had to code using assembly language or "machine code." Even with my numerically challenged brain, I found assembly language reasonably easy if I thought of it in terms of little boxes containing zeros and ones.

In the short span to two or three years, the computer games "cottage industry" metamorphosed into the "computer games industry." The hundreds of small games producers gave way to the megalithic few. Today, computer games are on a par with Hollywood movies in terms of production and advertising budgets.

Leaping forward ten years, we see the same model emerging in the seminal multimedia business. From early home-made HyperCard stacks in glorious black and white to current mega-produced CD-ROMs from Microsoft and Disney and all-singing, all-dancing, multi-screen corporate presentations, multimedia has matured from its initial gross overclaims and hyperbole to a solid and useful communicative medium.

My previous design and programming backgrounds made multimedia a natural working environment for me and my introduction to web design started as a multimedia project. Before the WWW came along, a major British newspaper commissioned me to do a presentation I called "Tomorrow's Newspaper".

It was all science fiction and demonstrated how a daily newspaper could evolve into an electronic one. After discussing the 'delivery' possibilities -- CD-ROM, Flash Cards, optical card, satellite, cellular radio -- I pointed out that the real benefit of such a newspaper, apart from saving trees, would be its interactivity.

I went through a process of examining the possibilities of delivering the newspaper via the telephone system and modems. It was clear that there had to be a host computer with a program that could produce layouts and squirt then down the wire and client programs that could decode them and present them faithfully on the user's machine. The time scale and cost of this exercise was a major drawback but another stroke of luck brought a cat and an early beta copy of NCSA Mosaic to my doorstep. It couldn't do all things we wanted, but it was available - free!

Within a few days, I had learned HTML and had a working mock-up of the newspaper running on my PowerBook and gasps of awe from the client. Within a few months, the "Electronic Telegraph" was up and running on the Web.

So, you could say that I got into web page design through having a vision or making a prediction.

Now we have the World Wide Web, and history repeats itself. The familiar signs are there, the cottage industry beginnings, the hype, the increasing sophistication, the big bucks. But the main difference between the Web and the games or multimedia industries is in the scale and diversity. Never before have so many people across the world been involved in such a concentrated effort. Never before has it been so accessible.

Where the technology works at a basic level at present, the variety of tools available to people creating web pages is multiplying and diversifying at an alarming rate. There are already more than eighty plug-ins available for Netscape Navigator on the Mac alone. What we have is a veritable Swiss Army Knife, a plethora of tools each of which do their own little task - and handy in an emergency situation - but not necessarily the best tools for the job.

Round the Bend

I think it is fair to say that any graphic designer coming into web page design from more traditional print areas will be frustrated with the lack of control offered by basic HTML.

HTML is not a page description language like PostScript or a programming language like BASIC or Pascal, it is somewhere in-between. It offers rudimentary layout possibilities somewhat akin to a word processor and the ability to link to other information - not much else.

The vast arsenal of bolt-ons and plug-ins available extend the possibilities considerably, but not consistently or reliably. You can either choose to ignore such extensions on the basis that most people will not be able to take advantage of them, or you can push the envelope so that an enlightened few can have an enriched experience.

The next generation of web page production is starting to address these shortcomings of layout and programmability. The main contenders in these areas are Cascading Style Sheets (CSS) and the Java programming language. And to tie these technologies together, we have XML (Extensible Mark-up Language), richer and more flexible than HTML.

Cascading Style Sheets are an extension of the HTML language. If you use style sheets in programs like Microsoft Word or Quark XPress, you will appreciate the benefits of being able to assign styles to headings, sub-heads, paragraphs and captions. If a page is designed for Cascading Style Sheets, browsers that support them will benefit in terms of layout, but non-savvy browsers will still work with a degraded yet acceptable presentation.

Sun Microsystems' Java language has made great strides in the last year. Its main benefits are elegance and portability. It is considerably easier to write (and read) than its main contender C++ and it will run on any computer that has a Java runtime compiler. But it is a young language and many issues have yet to be addressed in terms of functionality and speed. It does need a fairly highly specified computer to give acceptable performance.

Currently, some companies are trying hard to push Java as a web page layout mechanism. Using Java as a page description language is perfectly feasible and is being pioneered with products like UniQorn and WiredWrite. The Java applet still has to be downloaded and hosted from within an HTML page. Even with superior compression schemes, Java applets have to be downloaded completely before they can execute, which can make the whole process seem painfully slow.

XML provides a catalyst that marries style and function. Where HTML provides a set of pre-defined tags, XML lets you define your own, just as you can define your own style in CSS and your own classes or functions in Java. In fact, a tag can be a combination of a visual style and a programming function. The real benefit of all this is that web page documents will not only look better, they'll be more intelligent, too. They will edge closer to becoming virtual applications able to carry out spreadsheet and database functions in addition to the mere delivery of multimedia content.

The Crunch

Unfortunately there is a cost to having all these extra tools and facilities. Not so much an Internet bandwidth problem -- because connection speed is increasing all the time -- but a human brain bandwidth problem. The polarization I saw in the games and multimedia markets is happening again, where the whole process is becoming too complex for the 'cottage industry' web page designer. S/he will probably be content to stick to HTML for their personal home pages because it serves that purpose reasonably well, but at a professional level, web pages will become major production efforts requiring teams of specialists.

Any other communications industry has the same structure – be it print or film and television. Television has two main standards, 'home movies' and "broadcast" with very little in the middle. What little middle ground there is, is taken up by the corporate and educational sectors, both heavy users of web technologies. Although standards vary enormously in these areas, they aspire to the professional broadcast end, budgets permitting.

Home pages produced from home are not going to change drastically, but corporate sites will no longer be able to look like personal home pages.

The whole web delivery process is evolving and, like the Darwinian Theory, is subject to processes of natural selection. The useful tools and facilities are going from strength to strength, and the lesser ones falling by the wayside.

The one prediction that I can make with great certainty is that the future of the Web will never be predictable.

Joe Gillespie is a designer/art director based in London, England. From a background in leading advertising agencies and design studios, he now heads Pixel Productions, specializing in design for multimedia.

Joe has acted as an interface design consultant to Apple, Microsoft and Videologic in the UK and was involved in the conception of the Electronic Telegraph &emdash; one of the web's premier newspapers.

Currently working on CD-ROM projects for Canon and Sony, Joe has also produced the High Five award winning Web Page Design for Designers site.

Del.icio.us Digg Technorati Blinklist Furl reddit Design Float