The Semantic Web
Before getting into the Semantic Wusic project, an overview of the semantic web might be appropriate.
Getting computers to truly understand what humans create has been a holy grail of sorts ever since we started linking information together via the Web. It is one of the most stark differences between the way we think and the way computers “think”.
In the late 90s, a bunch of really smart PhD types got together at several conferences around the world and decided that the Web had some growing pains. For the Web to really shine, we needed to teach computers how to understand the body of human knowledge that was making its way onto the web. The idea of a semantic web was born. This new web would help computers understand that when we typed “Hung Up by Madonna” – we were talking about a song called “Hung Up” by the artist Madonna. To a human, it is clear that we are talking about the song “Hung Up” by “Madonna”. To a computer, “Hung Up by Madonna” it is just a random collection of letters and spaces – it doesn’t have meaning.
Computers are good at doing things that they are instructed to do, but have a hard time inferring relationships or understanding what humans talk about unless it is painfully spelled out to them. In short – the web became far too complex for a single person to truly understand. We needed computers to automatically organize the information for us. Unfortunately, computers don’t readily understand key concepts our writing, nor how those concepts relate to one another.
So the semantic web, at its core, is a way to get computers to understand what we write about in our websites, MySpace pages and blogs. If computers can understand what we write about, they can help us find what we want more easily.
But how do we teach computers what we’re talking about…