Serendipity35

Too many choices. The Net is even tougher than one of those big Jersey diner menus. I heard a librarian say once in a presentation that finding something on the Internet is like "trying to take a sip from a firehose." That's why we all need filters and recommendations.

This is hardly new. How do you choose what books to read, TV set to buy, and movie to go to or rent? And how did you do this before the Net?

The answers certainly overlap - professional reviewers, word of mouth, friends recommendations, sales figures, advertisements.

Example from the past weekend...
I like Robin Williams. He has a new film out called The Night Listener. Do I want to see it in the theater or just add it to my Netflix list and wait for the DVD?

So, I decide to take a few sips...

1. The film has the obligatory Flash-heavy website, which we know is highly prejudiced.
2. You canwatch the trailer in a number of places, including on YouTube
3. I can check reviews. My taste often matches up with Roger Ebert.
4. If I want more on the actors, I'll always check the Internet Movie Database.
5. When I check theater times in My Yahoo, there are 10 "pro" reviews of the film and reviews from over 600 users.
6. My wife says she wants to see the movie.

So what did I get from all that?

1. Nice site. Odd that they allow you to download a podcast (that's cool) but that, since it's a film, it's an audio-only MP3.
2. The trailer runs 2:27 and is intriquing.
3. Roger gives it *** out of 4 - a "B" on the ever-popular 4 point grading scale.
4. Has reviews by users but is more informational. I am reminded there that I like the film's co-stars (Toni Collette , Sandra Oh) and that Armistead Maupin wrote the original novel & collaborated on the screenplay, and where do I know the ex-boyfriend from -- ah yes, he was Will's boyfriend on Will & Grace. IMDB is a great argument settler when you're talking movies.
   
5. The pros average out to a "B-" but all of us have averaged student grades or had our grades averaged, so we know what that means. USA Today gave it an A- ("...features Robin Williams' most subtle and quirk-free performance since 2002's One Hour Photo. It's also his best.") and The New York Post gave it a "C-" ("...showcasing Robin Williams at his maudlin worst.")
6. I go to the movie with my wife. I give it a "B." My research was accurate.

Netflix and Amazon have recommendation services for movies, books, and music. They keep a profile of your purchases and ratings but the recommendations are pretty standard - titles by the same artist, author, director, or something in the same genre. I bought some kids DVDs, gourmet cookware & a bathroom scale as gifts from Amazon, so they went into my profile. So, I started getting suggestions for those things. You need to "revise" your Amazon profile so it reflects your real purchasing.

Netflix tells me that I have reviewed more than 1600 movie titles and that I have 500 films in my queue (I expect a long, happy retirement one day watching movies...), and yet they only have ONE recommendation for me. They tell me,"We can't find any more recommendations for you. To get more recommendations, rate more movies you've seen." Oh, come on - I think my review numbers are pretty robust already! Work on that algorithm!

Some online retailers use recommendation engines to suggest an additional related purchase. How about an extra battery & memory card for that digital camera you're buying? I find this helpful because they are likely to suggest the correct battery amongst the too many choices, and might even remind you of something you actually do want to add to the order.

Other sites, like CNET and ePinions, are focused on the user reviews and the links to retailers is the second step (though an important one for them as they get a slice of the sale). Am I the only conspiracy theorist that wonder when you see a glowing product review on one of these sites that it was written by the manufacturer?

There are also the social bookmarking sites that allow you to add your review or vote to sites and stories online so that others can see what the community thinks is good. Three popular ones are Digg, Reddit and TechMeme.

Finally, in some ways bloggers can act as editors/filters for information. If there's a blogger you trust to give good information, you use that blog to filter through information. For example, I like the choices that Brock Read makes in her Wired campus blog for The Chronicle of Higher Education. She gets paid to sift, read, and blog and I'm a subscriber, so why not take advantage (you can read it for free though).

Why not use these sites to mine the web for yourself? The search engines do it - using opinion leaders and authority figures to rank results. Google's ranking algorithms (PageRank) weighs heavily on measuring which pages have lots of inbound links. They're using the logic that if a lot of people link TO a site (especially if those links are themselves ranked highly), then it must be good.