Favorite5.com: Less is More

Welcome to Favorite5.com

I love to read. I love movies. I love music. I'll bet you do, too. I feel almost naked without a book in hand, or in my bag. I carry my entire music collection with me everywhere I go, on my laptop and iPod. And a weekend just isn't complete if I can't walk into the office on Monday, raving or ranting about the latest movie I just saw.

As a voracious member of the media consuming masses, I frequently run into a problem: What to read next. I walk into a book store, and there are hundreds of choices, but which are the good ones? I slowly scan the fiction shelves. Is this the next book that will change my life? Who knows? Somebody must.

In this technological age, we have turned to computers to help us sort out the almost oppressive mass of information. Maybe if I told a computer about all the books I've read, and what I thought of them, it could look at what other people think, and the computer will point me to greatness.

One night, a few months ago, I turned to Amazon.com's recommendation system. I had just visited the book store, unable to pick out a winner. Perhaps I could gain some insight. I've been an Amazon customer for a long time, so I figured they would have plenty of data on me.

I was sorely disappointed. As it turns out, I had recently purchased a book on CSS, which web designers use to make web sites look good. I rated it highly, because it was a good book. But I was on to other things. Amazon, however, proceeded to recommend seemingly every single CSS book in their catalog. I had to click through page after page. I walked away befuddled and disappointed. There had to be a better way. I didn't care about CSS. I cared about fiction. Wouldn't it be great if I could tell the system a couple of examples, and say "Give me more like these"?

The problem is not unique to Amazon. Indeed, most recommendation systems I've seen are lackluster. Computer systems have too much data. And the only way to work through such a crush is with taste. But computers don't have taste. All a computer sees is 1's and 0's.

I've had computers recommend many things I liked well enough. So-so movies. Unobjectionable music. But I've never had a computer recommend something transcendent. A computer has never told me about a book I just _had_ to read, right now, at all costs. But people have. So that's our plan. Instead of using computers to point you to books, we're using computers to point you to people.

At Favorite5, we're putting our trust in people. The key to finding good things to read, listen to, and watch is not punching a ton of numbers into a machine. Two books I've recently read that I really enjoyed were Carter Beats the Devil by Glenn David Gold, and The Time Traveler's Wife, by Audrey Niffenegger. Both were recommended to me by friends.

Of course, not every recommendation I've heard has been good. Some people do not share my tastes, which tend toward the eclectic. I knew a girl in high school who saw Titanic a dozen times. She was a lot of fun to sit next to in Calculus, but I suspect her advice would not be nearly as good as someone who, like me, believes Northern Exposure to be one of the greatest TV series ever produced.

It became clear to me that, if I was going to find great books, I was going to need to find great advisors. Thus was Favorite 5 born. The idea is simple. You tell us what your five favorite books, movies, and CDs are. We'll tell you whose tastes most closely match yours, and offer you the chance to connect to these kindred spirits. We'll show you what else they like. You can send messages to each other. You can make up a shopping list to print out, or if you prefer, hop right over to Amazon to make a purchase.

We're just getting started, and we have lots of ideas. If you have one, drop us a line. In the mean time, sign up for an account, and tell a friend. Somewhere out there is a group of people who know what you like. We want to help you find them.