I got my invitation to be a beta tester for [true knowledge], and I’ve been playing around with it. Basially, it’s a search engine that can handle natural language search queries, and find relevant information even if the specific terms used in the search query aren’t found. This is a fascinating technology that I wish I knew more about, but from what I understand, [true knowledge] makes use of existing databases of knowledge, like Wikipedia and the CIA Factbook, as well as searches of indexed web pages, and it searches through this data to find information that answers the user’s query. The engine attempts to manage this by interpreting or “understanding” the user’s search query.
So far, I’m very impressed. It handled the easy stuff well, and even if it fails, it does so gracefully (gives you relevant search results, sometimes takes a guess). Since it’s in private beta right now and you can’t go try it yourself, you’ll have to settle for this video:
I’ll write more about [true knowledge] once I’ve had more experience using it. For now, I’ve tried a bunch of easy search queries, and most of them have worked. Weirder phrasing makes it less likely the answer will be found, but you don’t usually have to bend over backwards to write it in an understandable manner.
That looks totally awesome! Does it really work?
Wow. This is so much better than Google. Whenever I wanted to find information about shitty pop stars on google, I had to read about them. I’m so glad that I can learn all about J-Lo now.
Seriously though, this technology is amazing. It’s always exciting seeing our knowledge of computing give computers get more advanced artificial intelligence. Although, I can predict a few downfalls with this system. I highly doubt it will be able to answer very specific and specialized or complex questions.