There were comments here, and on my ‘Apple Kill the Radio Star’ post on how useful the recommendations can be in addition to the star rating systems (like the “people who bought this also bought”). I did some digging, and there are companies that actually specialize on this for web-site back ends… these people could have a lot of influence!

Check out: http://www.loomia.com/

A.

I can think of at least one other drawback of Genius, Last.fm, etc. These services can easily become something of an echo chamber, whereby the listener consistently gets more of the same. If not the same song, the same style, and there is no built in mechanism to surprise the listener, no impetus for breaking out of habit. Mind you, conventional radio was transformed by the 1980s into a beast with the same tendencies, only magnified by very narrow financial concerns. This is part of why i stopped listening to radio the moment i started finding alternatives.

To avoid algorithms, one suggestion is, believe it or not, to use Amazon. This won’t save you from possible payola, but the music section at Amazon has a “People who bought this CD also bought…” feature. It doesn’t take long to get lost down a pretty interesting musical rabbit hole following some of those ‘also-boughts’. Of course, this is a little more time-consuming than an algorithm, but hence the reward, i’d say.

If you’re willing to spend even more time, there are sites like Pitchfork Media and the Internet Underground Music Archive. And of course, there’s always the blogs – most now have streaming song samples from the albums they review, and it doesn’t take long to find folks whose musical tastes overlap with your own, while also introducing you to stuff you wouldn’t even have thought to look for.

Happy hunting!

Enough of my favourite albums are poorly rated by “the people who take time to rate stuff online,” and I’m used to liking some things that most people just don’t.

I do often follow the “people who bought this also bought” links (and I’ve found some great stuff that way), but if I like the sample/description/hype of something, I’ll buy it regardless of the rating.

Your comments as well as a few on Facebook from Dave reminded me of a few services that actually showed a lot of promise, but have since disappointed.

Pandora, Last.fm & other internet radio stations that use smart algorithms to guess what things you might like are awesome. In particular, I liked Last.fm’s ability to dig up obscure stuff I could try. Unfortunately, our Canadian market seems to be locked out of some of these sites by copyright issues.

I just went back and had a look at Last.fm and I see the subscription is only $3 (USD) a month!! Based on that I think I will give it another try. Perhaps Apple doesn’t have the lock on this I thought!

I am also giving ‘MusicIP’ mixer a try as a ‘Genius’ alternative. So far I am not a big fan of the GUI, but it did seem to spend a lot of time analyzing my library, so we will see how smart it is.

Hype Machine looks very interesting.

Thanks for your comments, both on the blog and off!

A.