In Hot Pursuit of the Edge

Human interface design is one of those things I hardly get to write about, but it's been on my mind of late. It's come to my attention, particularly of late, that there's something definitely wrong with how people are forced to interact with the Web.

Hypertext Markup Language (HTML), the markup language that drives the web for the most part, was originally invented by Tim Berners-Lee ([1], [2]) in the early 90s as a method of disseminating documents that had cross-references in them among his colleagues. Since then it's been published as a standard and constantly developed and paired with other technologies like XML and CSS, as well as dynamic scripting languages like JavaScript. It is still, however, primarily a language for marking up text for easy display on the web, just as languages like LaTeX are used to mark up documents for printing ([3]).

I would say that for most of the content of the Internet, this is just fine. People can access most sites kind of like they would interact with a book or a magazine. Clicking on links is a shortcut to page-turning, and content is presented like print is. But why should the other percentage, which is growing larger day by day particularly when we hear all this talk of AJAX and Web 2.0 awesomeness, still have to behave like that? In just the same way that users form expectations of how applications behave (mac users like to see "Preferences" under the "Edit" or Apple menus; PC users like to see them hiding under "Tools") depending on their experience, the same goes for Internet users. Users really like to see their websites broken into sections, and these sections broken into sections possibly, and they like to see navigation options between them displayed at the top of the page. Users have grown to like multicolumn layouts, and they've grown to dislike framesets ([4], [5]). That's all well and good.

My question, right now, is this: why do we have to always put nav links at top, have little cross-links in the body, and have a page layout that expands vertically, with only one column of text? Why can't I have, say, a picture gallery that's 600 pixels tall and 1300 pixels wide, scrolling from left to right? For me, that would be more intuitive. You're then interacting with a web page like you "interact" with the walls of an art gallery. Your mouse's position can tell the application where you want to go and what you want to look like, perhaps with a smaller version of it down below, scaled to view, so that the user can skip around via thumbnail links. If you're looking at a chart with relational data or a network topology displayed on it, why shouldn't you be able to click and drag the elements around to get a better view of them? Why can't you zoom in for more detail directly there on the page without having to go to another page that displays details?

You can. The only thing holding everyone back is that nobody's done any of this before. If your new system is intuitive and beautiful, if it's captivating and keeps people interested, then why not try it? The problem with interface design on the Web today is that nobody's trying anything, and it's really starting to bug me. This is not at all to say that nobody anywhere is trying anything at all to push the envelope. I'm just saying that I don't think anyone's trying anything that's truly unique to the Web, any kinds of interaction you can't find elsewhere in a better form.

I want to take the web and do something with it that nobody's thought of doing before, presented in a way that no one has ever seen but that they'll understand immediately. It should beckon users further into the interface without having some kind of annoying guide (Clippy, anyone? [6]). It should encourage users to play around and see what kinds of features they can discover on their own. A good interface should also prioritize so that features users want should never be more than two clicks away. It needs to retain the consistence of other user interfaces while at the same time increasing, overall, the variety of interactions possible.

That's all I've got for now. Give me a couple of weeks and then I'll show how I put all this into practice. Later, space cadets.