MediaVision Blog : Search Engine Marketing

I’m going to run the gauntlet and talk about the Semantic Web, and why it’s supposed to be hot stuff for SEOs in 2008. What I’m not going to do is hang myself first by explaining the new consumer applications that will herald its coming. From what I understand, this software is intended to improve the way machines review/connect/store/share/reuse data. It is hoped that they will clear the wood of the trees, or more literally, clear the Internet of locked data silos, so that we may apply our human creativity without getting wrapped up in the interoperability (or lack thereof) of necessary data. I guess what this means is that we’ll be able to share things between programs efficiently, which will be nice.

They say you never really learn how to swim until you get your face wet. For example, how can you evaluate the usability of an AdWords campaign without actually running one? I could go off down the academic rabbit hole and discuss annotating asymmetrical indexes to make them machine readable. But I’ve never actually encountered one, so there isn’t much point. I could dream about the Wonderland of equivalent data, from which the geeks hope to create generic and vertical search, contextual shortcuts and previews, personal information management systems, semantic browsing tools and a nice little machine that will read my thoughts and make me toast. All of this is what is foreseen in the coming of the Semantic Web.

But something about it makes me uneasy.

I worry about this fervour to classify things. In some instances it feels like a holy quest to improve the world’s intelligence (and its capacity for intelligence), simply by arranging knowledge and information in a way that complements that knowledge and information. I like to say (smugly) that a glut of knowledge creates a dearth of wisdom – so now they’re trying to manufacture wisdom. What we’ll get with Web 2.0 and Semantics is improved findability, automation of complex searches and dynamic web browsing. But I don’t know if we’ll really get intelligence.

Is the Semantic Web perhaps a little bit… hopeful? Let’s get back down to basics. Semantics, very basically, is the study of meaning in words. Data is classified using words. Assume that the words people use in searches link back to their word-object associations. And that in different countries these words differ depending on the associations specific to each culture and language. To extrapolate, the Semantic Web is implying that the long-term future of search could be arranged not by what Google sees as “categories”, but by how the same words in different languages have evolved to mean different things. And let’s not forget that it took the entire history of civilisation to make these innumerable associations. Thought leaders hope to categorise the web (our largest portal of shared information) so that literate people can get smarter, in any language. Honestly, if I keep typing I might break my italics key.

I feel reactionary treading on the Semantic Web this way, especially as it’s no more than a baby in a cradle; I’m stealing its candy. If you have any counter arguments or cautionary tangents to offer me, please do. I’d like nothing more than to look to the future of search with genuine anticipation.