Google is reportedly testing a new feature in the Knowledge Graph sidebar: the ability for users to find out more about the topic and suggested results. An “Explore More” option has appeared to some users, allowing for an expanded set of results below the search box.
Google Operating System posted the above screenshot of the test in action. Google is constantly performing tests in search results, so this one should come as no surprise.
It is interesting, however, to see what they might have in mind for Knowledge Graph. When it first launched mid-May, Google called it “the first step in the next generation of search.”
Knowledge Graph made sense of the 2010 acquisition of open content aggregator Metaweb and their Freebase, a database of more than 12 million entities. Freebase isn’t Knowledge Graph, though; it’s just one cog in the massive machine that powers Google’s instant-answers-in-the-SERPs for informational queries about all manner of people, places, and things.
As of May 16th, the Knowledge Graph database contained more than 500 million objects, with over 3.5 billion facts about – and relationships between – those objects. Other sources of data include Wikipedia and the CIA World Factbook.
With this test, Google is showing an intent for Knowledge Graph as a discovery tool, which would be about the most philanthropic and least evil purpose for it.
Some aren’t so sure about Google’s good intent: WordStream threw out a conspiracy theory earlier this month, questioning whether the purpose of the info boxes is to get users accustomed to looking to the right of the organic results for information. (She has a solid point, conspiracy theory or no.)
Of course, this test could also be used simply to gather data on the relationships between objects, allowing Google to improve results on regular Knowledge Graph-applicable queries. To dismiss any Google feature without considering its financial implications, though, would be naive.
Google described the next generation of search as tapping into “the collective intelligence of the web and understanding the world a bit more like people do.” Collective intelligence and drawing from wikis or open source has its drawbacks, though; we saw a few weeks ago that Knowledge Graph results have serious lag time when bringing back queries that should show recent events or updates.
Whatever its purpose, tests like these are fun to spot, even if they only lead to more guessing as to what Google might throw our way next.