Anne Truitt Zelenka

Adam Green calls James Corbett’s Web 3.0 wonderings “snark bait.” Here’s some more chum—bring on the snarks. I’ve been thinking about what Web 3.0 will bring for two months now. Two months ago I came back to tech after a five-year stop out and realized that the Web had shifted paradigms while I was changing my kids’ diapers and begging my doctor for Prozac.

If you dislike the idea of Web 2.0, if the term makes you cringe, if you think it’s all hype, STOP. Don’t read any more. This essay will only annoy you. I’m sure you have other things you could read instead. I find Web 2.0 a convenient shorthand for how Memeorandum differs from iVillage, two websites that to me represent the second generation and first generation of the Web, respectively. But I don’t kid myself by thinking that there’s some strict dividing line in time or in feature set. It’s a convenient fiction, that’s all. Those who get worked up about it may be taking it too seriously, as though there are only two choices about Web 2.0: either it’s a Biblical truth or it’s completely false. Me, I take a Buddhist approach. I accept Web 2.0 as tentatively and pragmatically true, not as a statement about objective reality.

James offers a table of how he sees the distinguishing characteristics between three generations on the Web. I mostly agree with how he lays it out, but I’m thinking that the unit of content for Web 3.0 is not an RSS article but rather a more granular unit of information as could be represented by a URI in an RDF triple. I’m thinking that RDF may be the key Web 3 technology rather than OPML. Anyway, those two technologies are not mutually exclusive or replacements for one another.

Here’s how I see the evolving web:

• In Web 1, the basic Web element is an HTML page. Presentation is mixed up with what little semantic information exists. The browser is the key way of viewing and interacting with a page, but it’s mostly about viewing, not interacting, except when the user submits information to some sort of authority via a form. This is the browsable web.
• In Web 2, the basic Web element is an RSS article or a page structured as XHTML/CSS. Now it’s the read/write web. Users mainly interact using news readers (e.g., Feed Demon or even Memeorandum, a cell phone or PDA) and news producers (i.e. blogging engines, content management systems).
• In Web 3, the basic Web element is more granular. I don’t know what to call the individual elements, but each has its own URI. This is the semantic web, the read/write/process web, the active web. The web will “know” so much of itself via semantic technologies like RDF that tasks that previously required human input will be able to be handled mostly by computer algorithm. The web’s greater level of understanding means that users can go further than publishing their thoughts, they can create programs that help them achieve broader goals. Better yet, the active web itself can create and run programs to help users achieve their goals online.

If I sound hazy on that last one, it’s because I am. I’m still puzzling out what the semantic web gives us and what it means for social and technological interaction on the Web. But I feel sure that it has the potential to overthrow the current paradigm.

UPDATE: In his comment on James’ post, Danny Ayers offers a helpful example of what I’m thinking of as the active web:

As a example, consider calendaring. I could put my schedule online in HTML, you could do the same with yours. A human looking at those might see “they’re both going to be in London in June, maybe they could meet”. But if this information was made available as machine-readable data, a computer could make that inference and tell us both.