h/t CM, Intro GIS
Last.fm is a free online service that offers its users the ability to both catalogue and track their current music listening habits as well as to expand their tastes by creating personalized recommendations and streaming radio stations based on aggregated data of user preferences. For example, the algorithms are able to match you with “neighbours” who have similar musical tastes and make suggestions based on artists you haven’t listened to but your neighbours have. The accuracy improves as you listen to more songs and log them in your profile. The success of last.fm (it has millions of users in almost every country in the world) has enabled it to track broad trends in music listening worldwide. Weekly charts are compiled that show the top songs and artists overall, as well as the fastest-gaining tracks and musicians.
Last.fm has made a recent foray into the world of iPhones, in which users gain access to their own personal radio stations while on the move anywhere in the world. This application has made it clear that there is much potential both to expand the applications of this technology. Tim Walker, writing in The Independent about the possibilities of moving data storage online into a “Cloud” says that, “If the iPhone and Google-phone allow their owners to listen not only to their own music libraries but to any music at all via the Cloud, they will quickly make that old-fashioned data-storing device, the iPod, redundant.” I offer this anecdote only to demonstrate the potential for massive amounts of data to be collected and shared about music listening habits with relative ease based on existing technology. What is most fascinating, and has not yet been discussed, is the potential for Last.fm to make charts like the Billboard Hot 100 irrelevant.
Billboard’s decision to include online music sales in the formulation of charts shows that they’re aware that the future of music can be found on the Internet. But if the vast majority of music is being either downloaded illegally or streamed legally from sites like Last.fm (or potentially from other networked users), then a lot of real life listening is going unrecorded and is thus not reflected in what is defined as “popular”. Last.fm could make popularity a lot more democratic than it is now, with radio play and album sales largely determining chart placements and a spot in the cultural canon. Not only could Last.fm lay bare the hidden epidemiological workings of how music trends take off, it could actually modify how those trends work. If users have access to millions of songs, they’ll probably start by listening to what’s already popular or on the rise. Thus something could become a hit due to the snowball effect of online users selecting it without any marketing campaign whatsoever.
What does this have to do with GIS? Last.fm currently asks users only to identify their country and time zone. If it asked for more detailed information, or merely used IP addresses to determine location, and encoded every track submitted to the database with this information, the potential for customized charts is limitless. Charts could be compiled detailing the top artists and songs at the level of neighbourhood, city, province or country. You could ask for the top artists among users living within a 50km radius of yourself. In advance of moving to a new town, anxious teenagers could find out which bands everyone at their new high school was listening to before they even arrived. Obviously, this information would be highly sought after amongst marketers of just about everything: musical taste has a well-documented relationship with the consumption of style and accessories. Spatial location is only one type of data that could be linked to music listening – any demographic trait would work equally well. Charts could be created based on age, gender, income etc.
Last.fm has already begun to make some spatial connections with user data through its “Events” pages. Users can indicate which concerts they plan on attending by selecting their nearest city, and Last.fm can notify you if a band in your “library” will be playing near you in the future. This could be expanded to include record stores and other music-related business ventures.
The level of public availability of this data will therefore be of huge importance. If Last.fm, owned by CBS Television, chooses to make everything public (while maintaining user privacy), this could be of great interest to everyone from sociologists studying social trends and cultural capital to musicians interested in genre and audience reception. If Last.fm releases the data selectively or at cost, it will serve the interests of record labels, radio conglomerates and concert promoters attempting to find profits on the internet to replace stagnating revenue streams in their traditional markets.
Links:
Last.fm now enables streaming of Billboard hits. This is the bare minimum of the potential offered by this technology, which could rewrite how Billboard hits are themselves calculated.
Although not spatial in nature, this map of Last.fm artists demonstrates the potential for drawing connections between artists and listeners based on user-generated tags and listening statistics.
A look at how people are willing to pay for the “right” music at the right time and in the right setting. Last.fm essentially offers this service for free already – listeners can select from user-generated tags and stream a radio station of music tagged “relaxing” or “playful”. Adding spatial data to last.fm could enable the option of radio stations linked to particular places or atmospheres, like music popular in the Caribbean or the Rockies.
An overview of the “Cloud” and how Last.fm could gain popularity under this system due to availability on wireless devices.