MediaVision Blog : Search Engine Marketing

To understand the semantic web is to understand the future. I’ve attempted to begin this knowledge quest practically in reading about Yahoo!’s latest developments in what they call their contribution to the semantic web. ”Open Search” is Yahoo!’s endeavour to enrich search results with deeper information, where integrating the support of semantic web standards such as microformats and Resource Description Framework (RDF) into search results, is the way to do so.

Standardising these meta data formats will allow website owners and developers the means to make better detailed site topic or subject matter available to the search engine to include in search results. In this way, the semantic web will “open” search (giving site authors more control over what to include in the results).

I find this step, while acknowledging that it’s probably designed with significant engineering prowess, a rather feeble and disappointing outreach to gain the attentions and affections of both searchers and site authors. It’s too little too late, I’d say. In the face of Webmaster Tools, Universal Search and the massive relevance advantage of Google’s powerful search platform, and even Microsoft Live’s Webmaster tools version, do Yahoo! really think such baby steps will affect real sway in the search race? Perhaps they have some interesting developments in the pipeline that depend upon adding webmaster-controlled data into search results; but time will tell.

Google’s spam team often beckon webmasters to develop for “the user”, rather than to appease the search engine, but this is because the indexing bots and the algorithm are not water tight in their efficacy, and on their own are inadequate to fulfil the vision of a bias-free search experience for all. The process, in all its complexity, still cannot distinguish independently the many gaming and spamming tactics from the “legitimate web”.

I’d like to see the single search query system deliver not vast amounts of relevant results, but rather a few very relevant ones. Never mind the company contact details and such that Yahoo! insist will enrich their results, and forget the query suggestion tool that pops up at the Yahoo! search prompt while you punch in your query.

Semantics (not ”the semantic web”) dictate that words are signs that signify an object. So words or phrases by themselves are useless to a person searching the web. The meaning behind the word is the important bit that the searcher should ultimately arrive at. It’s great how Yahoo! are trying to fulfil Tim Berners-Lee’s vision of a semantic automated web, where “computers become capable of analysing all the data on the Web – the content, links, and transactions between people and computers”, by fattening their results with volumes of related data for searcher satisfaction. It would be greater still if the results were less confusingly crowded (with the exception of results from excessively long-tail queries).

It will be satisfying when the search engine can do a big portion of the thinking for you, somehow mind-reading your needs, so you don’t have to articulate your intentions to the point of requiring a PhD qualification to do so. What is disappointing is that all search engines can really do to gratify our needs is to guesstimate which web pages contain what you want and fire buck-shot at you hoping something will hit, which works fine in some cases but certainly not always.

I moan bitterly, only to express my need for a better experience, where I can peacefully search without accommodation sites popping up while looking for television sets, for example. Maybe Yahoo!’s compliance with a standard will chisel new contours into the search landscape, where once again webmasters, rather than search engines, will sculpt the web.