Spying on Big Brother

I have long been fascinated by the concept of metadata. Data about data – or information about information. Unknowingly we are leaving behind a breadcrumb trail of metadata through our behavior, likes and dislikes. With the help of today’s fascinating search technology we should be able to discover answers to questions we have never even thought of asking.

What is it that makes you like something? We hardly ever stop to think about this. If only we could break that something down into small elements and study the attributes of every element then we would probably be able to spot a pattern and find more things we like.

The Music Genome Project shows us what is possible. The team claims that they have analyzed more than 10,000 songs and recorded information about every imaginable attribute that defines musical quality. This clearly goes way beyond genre, rhythm, or “customers who bought this also bought that”. (If you have not done so yet, discover new music based on your likes and dislikes at Pandora.com).

It should however not be necessary for a team of people to painstakingly dissect and tag attributes. Google should be doing this for us by identifying patterns. If I could list a few artists and songs I like, and a few that I dislike – why should it not be possible for Google to find similar likes and dislikes and spew out a list of the top-10 songs I should have?