We Are Web Technology

Recently, I had an idea for a classy text interface that I thought would revolutionize how people used the Internet. Within five minutes of online research, I discovered that someone else had come up with the same idea... in 1986.

The lure of Web 2.0, a.k.a user-centered computing mechanics, is finally beginning to outstrip the humble origins of HTML and its basis in a formatting language for articles and other educational ephemera. But we have many decades-worth of ideas within the coffers of both academia and popular business related to interface design just yearning to step onto the stage of the Web. And I'll wager that it will take another two to three decades to reach the point where Internet users are comfortable with these advances. They will revolutionize how we engage with and consume information.

The initial beauty of hypertext was its ability to channel and shape large volumes of information and allow linking to similar types of content. The real revolution to occur in user-centered Web systems has little to do with typesetting content, and everything to do with contextual organization and passing knowledge through conversation.

Let's Create a Little Context

I want a fluid markup language that is easily usable by the layperson that automatically allows us to build intelligent, intuitive relationships between different types of content. I'm not talking about XML, per se. XML is the enabling factor that will encapsulate all Web content. I'm talking about the browsers that parse XML and spit out the Web pages we love so much in fresher, more usable ways.

For example: In the future, and hopefully the near future, I would hope to see Web pages and/or browsers where, if you are reading on a topic of interest, the article could reconfigure based on areas that you linger upon, offering up topics that may be of interest in a separate sidebar.

Or, if you click on a link within the paragraph, instead of being tossed into another browser window or tab, a secondary channel on the page opens to enclose and accommodate that content, allowing you to choose a path of inquiry in-line before punting to another location.

This would require sifting and sorting of content elements that goes beyond what we call content management. It would require a method of relational sorting and artificial intelligence that may not exist yet in the consumer space, though I have a feeling it already exists in some form at Xerox PARC or Apple, but is being withheld from the general market as it is not seen as "usable" yet.

DHTML and AJAX are allowing us to add another layer of context within Web pages, so we can explore deeper levels of information within a single Web page. And yet it's still not enough, and that kind of functionality is usually custom-built at great expense. We're still bookmarking pages (socially or otherwise) and leafing through the online equivalent of a magazine. A well-designed magazine that evolves over time in response to my activity, mind you, but still a magazine. Most related links are hand-coded in, and companies like Amazon and Netflix have invested millions in creating automated systems that create context. Bring their systems to the masses. Organize us relationally.

Either that, or rely more heavily on the content management system that we call The Human Brain.

That's What I'm Talking About

What stands in the way of this contextual method of organizing and navigating Web content? Ourselves. Humans have yet to evolve to the point that they can move beyond the static page en masse. Perhaps the next generation of Web users will be able to grok this method of information navigation without feeling like the ground is about to give way beneath their feet.

It's clear that the Internet has turned into the Mariana Trench of human knowledge. Trawling that almost infinite depth of information has become entirely impersonal. We don't control how we reach the things that we need, and we can't easily control how that information is presented. When couched in these terms, it's easy to understand why it's so hard to create a Web system that users would want to experience.

Novels, essays, articles, blog posts, diaries, text messages -- these are all fixed forms that govern how we apprehend content and structure how we consume them. When we compare the longevity of those forms to a fluid content bucket that can't be passed along or referred to another person without changing in some way, you can only imagine the fear it would strike in an everyday user. Just give me a page with what I want, or a search engine that can find me what I want within two clicks (unless I'm feeling lucky). Seeking knowledge, fulfilling task, receiving information, end stop.

Interestingly, if you think about how people seek knowledge independent of the Internet -- through real-life queries and conversation with peers that have reservoirs of knowledge -- then you'd think we'd be spending more time creating Web systems that mimic that human-to-human mode of information acquisition, that add shape and color around information that we seek. I see this as not Web 3.0, or even 5.0. Until there are heuristics that can resolve these levels of complexity in a piece of content, and spit out a relatively appropriate answer the majority of the time, this technology will be a novelty used by dozens.

Thankfully, we have the time and the depth of material out on the Internet to use as our laboratory, as well as the new explosion of smart devices and micro-computers that will worm their way into the fabric of our everyday activities. By the time this type of contextual navigation reaches the masses, we'll be so plugged in that we'll probably just be querying people instead of computers to receive what we need. Oh, wait -- I just got a text message from a friend asking me if I know a good place to find high-quality dark chocolate...

[UPDATE: Check out powerset.com for a taste of what I'm talking about... currently it works with a subset of information-dense sites like Wikipedia.]