blog.rightreading.com

Somehow I seem to have missed the news about Google Sets. Google Sets are G’s way of categorizing the web (and therefore the .. well, everything). It is, in other words, computerized ontology (ontology being the study of the broadest range of categories of existence).

The way this works in practice is you enter one or more terms and Google Sets will predict how the set should finish. For example, if you enter “Asian Art Museum” and “De Young Museum” GS produces a pretty reasonable predicted set: de Young Museum, Asian Art Museum, San Francisco Museum of Modern Art, California Academy of Sciences, Exploratorium, Oakland Museum of California, Yerba Buena Center for the Arts, Fine Arts Museums of San Francisco, Legion of Honor, San Jose Museum of Art, Golden Gate Railroad Museum, Asian Art Museum of San Francisco, MH de Young Memorial Museum, Lawrence Hall of Science, Ansel Adams Center, Ripley’s Believe It Or Not Museum, Lindsay Wildlife Museum Walnut Creek, Berkeley Art Museum, Cable Car Museum, Laserium, San Francisco Maritime National Historic Park, Cartoon Art Museum, Hobnob Walking Tour of Nob Hill, Zeum, Breakfast following each night s stay, California Historical Society, Parking at The Fairmont Hotel, Crocker Art Museum, Fine Arts Museum of San Francisco, Accommodations at The Fairmont Hotel, Monterey Bay Aquarium.

Other times the results are … peculiar. A set begun with just “Maya,” for example, wanders off to produce the terms Manticore, Softimage, RuffStuff, Houdini, Lightwave, Photoshop, and Renderman. No doubt there are reasons for this. But they elude me.

Adding another level, Philipp Lenssen has recently produced a tool that combines Google images with Google sets, so that you can find related images. (Works with videos too.) I’m not sure how practical this is, but it’s interesting.