Music on Rails

Napster has a webservice that seems to be unpublished, but it is used in their website to get information on artists, albums and tracks. This is what I have discovered about the API which I use for Song-list.net.

I’ll detail the query language and not the results as they are in XML and self documenting. The data provided is quite rich.

• Artist: http://www.napster.com/searchXML/artist/herbie hancock
• Album: http://www.napster.com/searchXML/album/road escondido
• Track: http://www.napster.com/searchXML/track/beautiful love

Artists, albums and tracks all have 8 digit IDs that can be used to query the particular item.

• Basic information by ID:
• http://www.napster.com/searchXML/artist/10429542
• http://www.napster.com/searchXML/album/10499404
• http://www.napster.com/searchXML/track/10063400

• All albums or tracks by Herbie Hancock:
• http://www.napster.com/searchXML/artist/10429542/albums
• http://www.napster.com/searchXML/artist/10429542/tracks

• All tracks on “Thriller”:
• http://www.napster.com/searchXML/album/10499404/tracks

You can use the the informatin gathered to play the music through a browser:

• http://www.napster.com/player/tracks/10063400

Let me know if you discover more features.

This blog is for writing about song-list.net, and the technology used to build it. I’ve enjoyed lots of information people have posted online. Hopefully I can contribute a bit myself.

song-list.net is a side project (a distraction from) my main project which is expanding the capabilities of The Real Book Listening Guide (But that is a topic for another post).

The website was written using Ruby On Rails , MySQL and use of Amazon, Rhapsody (published), and Napster (unpublished) Webservices. Data from iTunes and eMusic was screen-scraped. In the upcoming weeks I hope to document some experiences using these tools.

The data collected so far is minimal (about 100 artists) since I wanted to go live with the application and see if it sparks some interest. Collection of data is automatic, but does require a human eye because all data is dirty. Over the next few months I should be able to improve the tools, the quality and the quantity of the data.