Of course, very few people are actually extremists. Structuralists accept that good presentation can enhance information, and presentationalists likewise accept that some structured information can enhance experience.
Since HTML first became mainstream, with HTML version 2.0, there has been a struggle between the structure of a document and its presentation. This battle is symptomatic of two competing visions for the web.
One web, two visions
Structuralists see the web as a vehicle for information: the web is a tool to simplify the lives of over-stressed human beings.
Structuralists believe that the presentation of a document is best left up to the individual user, browser or device. To them, HTML is a language for describing the structure of a document, and they view the future web as a universal any-device, any-where, any-time interface to databases, or as the database itself.
Presentationalists, on the other hand, believe that the web is for conveying experience – a medium for entertainment and artistic enrichment.
Because presentationalists want perfect control over their narratives, they think that HTML should be a language that can accurately present the web designer’s vision pixel-for-pixel, microsecond-to-microsecond, and interaction-to-interaction. The presentationalist sees the future web as a rich form of something akin to interactive super-television, but better.
Stuck in the middle with you
Most of us are realists who see the debate as largely academic until technology has caught up with theory. We are the would-be structuralists who still do page layouts with
<table> cells, and the rather-be presentationalists who still type out text instead of wrapping it up in images. The primary justification for most of our design choices is that “it works now.”
So what does work now?
Neither the structuralists nor presentationalists are winning outright. An examination of the latest slew of recommendations coming out of W3C might make you think that the structuralists are winning, but realists will quickly point out that technology has not yet caught up with presentationalists’ requirements.
Lone HTML (as it currently stands) is neither a good presentation language nor a good structural language, but the addition of XML and other highly structural languages and specifications is helping close the gap between reality and the structuralist’s ideal.
Cascading Style Sheets have promised and delivered major advances in presentation, while allowing document structure to remain intact. (The success of CSS is proof that time and time again we get the best results when structuralists and presentationalists work together.)
Good experience is hard to find
The web is not currently a good medium for experience.
In fact, it’s fairly cheesy when compared to the movie theater, the TV screen or even a piece of paper. Low bandwidth, differing interpretations of HTML recommendations, and a wide variety of user and device capabilities make it difficult for designers to exert the level of control necessary to evoke a consistent emotional response.
Not Just Slow Net Access
There are also cultural barriers to conveying experience on the web. We’ve been serializing narratives for millennia now: novels, movies, and television all rely on the user experiencing their narratives sequentially. The buildup of tension, suspense, grief, and other emotions takes time. Novelists and filmmakers have honed their skills in sequential media to a fine art, and users have become sophisticated consumers of sequential narrative.
By contrast, we are only just discovering ways of authoring experience and emotion in situations where the user is no longer passive, but is instead an active director of the experience. On the web, narrative has become a collaboration between the author and the user.
To be fair, live entertainment does already involve low-level collaboration between users and authors. A good storyteller, singer or comedian giving a live performance can be less sequential since he or she has the luxury of being able to evaluate the audience’s reaction and adjust the performance accordingly – something that movies and books cannot do. Feedback mechanisms do exist on the web, but are currently quite primitive in comparison to the interaction between live human performers and their audiences.
The current web also forces decisions – users must interact for the narrative to progress because nothing new arrives until the user clicks another link. It is currently difficult to author web content where the user can sit and experience a sequential flow of experience, only interacting when they choose to.
Interactive television has been discussed as a potential vehicle for the experience web; it combines the existing experience-conveying qualities of television with the ability to let users direct the narrative by choosing their own paths.
Even television, however, is a very limited output device that includes only sound and 2d (simulating 3d) vision. A truly experience-centered web could include many other output devices – imagine a horror “website” in which the site had full control over all of the speakers, holographic projectors, the lights, the doors, the climate controls, and even the smells within your house!
Television is also a passive medium. Its users have (and expect) very little control over what they see. The web is an active medium; users actively choose their own paths. Interactive Television encourages activity in a traditionally passive medium. The experience web may encourage passivity in a traditionally active medium. These two are already meeting in a middle ground.