Jennifer Zaino
SemanticWeb.com Contributor

Search leader Google has added a host of new features to its leading platform -- and support for semantic web technologies is among them. It is supporting the open standards RDFa and microformats as part of its new Rich Snippets feature for annotating meaning within search results.

Calling what Google does a "quiet rocket science," vp of engineering Udi Manbar at yesterday's Searchology event discussed the relationship between people, search, and Google. "We have many users and we have to solve their problem," he said. "If users can't spell, it's our problem. If they don't know how to form the query, it's our problem. Even if there's not enough content, it's our problem. If content is there but in a different language they don't speak, it's our problem. Even if the web is too slow, it's our problem."

The problems of search have changed from decades ago, such as getting enough storage, and now the focus can be concentrating on subtle understandings.

"Every improvement we make to search allows us to do more for our users, but at the same time their demands and expectations of us increase," said Gabriel Stricker, director of search communications. "Our fundamental task is to present to them the complexity of the web ... in a way that continues to be elegantly simple and straightforward."

So, what are some of the really hard, unsolved problems in search? There are five, according to Marissa Mayer, VP search and user experience, introduced a few "new shapes and sizes" to search: finding the most recent information; expressing that you want just one type of result; assessing which results are best; knowing what you're looking for; and expressing searches in keywords.

Approaching the whole problem begins with the new Google Search Options, which adds a "show options" link unveiling a new search options panel that lets you drill into different genres, different elements of time, and more. With the Recent Results option, a different algorithm combines relevancy and recency from web pages, news and blogs to deliver results; and users also can narrow results to a specific time, like within the last week, or to a specific category, like reviews.

You can also combine various options to customize results, building the query from the ground up so to speak -- for example, calling for images to be extracted from within the last week to add a visual overview to the text results. It also has added "sentiment analysis" which attempts to extract the sentiment of a product reviewer, which adds up to showing a different snippet than normally would have been shown in the results -- for example, the virtues of a solar cooker as extolled by a reviewer rather than perhaps the factual information absent of opinion that you might otherwise get.
Wonder Wheel is a different way to visualize results. Your query appears in the center and to one side directly related links, but to the other spatially related topics that you can drill ever deeper into. So you might work your way from a query about solar ovens to parabolic solar cookers to a brand new set of results featuring parabolic focal points -- and before you know it you're learning equations about parabolas.

From the circular wonder wheel to Google's next shape, squares: The new Google Squared will be available on Google Labs later this month. It's aiming at automated data extraction from the unstructured web -- delivering data in spreadsheet like rows and columns that was entirely populated from Google search results. To these users can add their own values that don't come up in the initial squares, such as a new category of dog to a search on "small dogs," or additional facts around the small dog breeds you're interested in, such as their energy level. But Google acknowledges that the Internet is still kind of a messy place, so automatically extracting value from there can have its hiccups -- for instance, squash shows up as a sports link under a Google squared table of vegetables.

With Rich Snippets, preview text for Google search results grow richer by the inclusion of new metadata that can make it more efficient to search for reviews, people, and the like. For example, users searching for a particular person can get better insight into facts about that individual to ensure they click on the right link the first time - among a host of Mary Smiths, for instance, you can key in on the one that works in San Francisco, for a particular company when the appropriate metadata has been tied to her professional profile on a directory like Linked In.

Webmasters who begin using RDFa or microformats to add these tags are not only going to help Google improve its search results, the company says, but are taking a big step to making the Internet smarter.