‘last.fm’, Mainstreaming the Long Tail?
Critical Analysis of: Last.FM web 2.0
Last.FM is a radiostation that scrobbles your music libraries on your computer (such as iTunes and windows media player for example). It takes all the metadata, like title, name of the musician, ratings etc and deduces your musicprofile from this. This way Last.FM can recommend to you the data of others with similar taste and start a radiostation with songs you probably will like that are shared by others in the Last.FM community.
The most important features of Last.FM are that it is easy to use and the speed with which you can use it (quick download, quick scrobbling of your data, quick access to a radiostation). But the social capital to be gained is seductive as well, you can join neighbourhoods and make friends. Last.FM profiles itself as an application by and for users and seemingly generates a lot of agency for every individual user, you can tag, love or hate, add friends etc. The exact amount of influence you actually have on the way you want to be recommended is blurred. If you have a specific track you want to hear you can not find this very easily and when you find it, it will only play for 30 seconds. There is not a easy way to go about to buy a song (as is the case with MOG.com which is a much more commercial web2.0 radioservice) via linking to commercial musicsites. In this system all users are equal and Last.FM is actually linking social capital, you can easily be attached to Prime Ministers Balkenendes neighbourhood if you share his taste of music. The commercial interests are not very apparent but under the surface you are giving away a lot of important information that can be used for marketing and the networkprofiles are becoming very clear, people of this age, in this country and so on are liking this sort of music…
On the blog on Last.FM you can access the Mainstream-O-Meter which is interesting if you consider the long tail (Anderson) theory. It becomes a ‘mainstream’ interest to have an obscure music interest to end up high in the ranking. This is making the long tail more visible but also more questionable. You don’t need to like a niche, you just have to find a niche and take it out in the open to appear in this ‘meter’. So really bad music possibly gets exposure because of the hype of the long tail effect. So then it’s more about the individual wanting to distinguish himself from the crowd then promoting a really good but non-mainstream article…
The tagging principle on the other hand is not really promoting the long tail effect because it only shows many used tags. So my tag ‘old folks rock’ for ‘the eagles’ is not visible in the tagcloud but only in my profile. While maybe I want to find more music related to this tag which I find really logical. With tagging you can easily make mistakes or even make mistakes on purpose to fuck the system up (Paris Hilton massively tagged as brutal heavy metal).
My biggest question remaining is on what the relationship with me, my friends and neighbourhood really is based. I can influence this only very indirect and the ‘how to’ is obsolete. I can not add friends of my own to my neighbourhood of taste or easily visit other neighbourhoods which limits my musical taste.
Fig. 1
Last.fm Mainstream-O-Meter, Bron: http://blog.last.fm/category/Found-On-Lastfm/
There already are a couple of approaches floating around the community to calculate your ‘mainstreamness’ manually. Here’s a well-designed web application by Luce. that does it for you: the Last.fm Mainstream-O-Meter. After Muz posted it in our internal IRC channel ages ago everybody went crazy for a while…
My score: 11.41% mainstream. So proud…
But of course you can always go lower. We already found out with the Chart Arcs experiment that Anil (aka joanofarctan) is a man of most obscure taste. I just found this comment from Luce. in the Mainstream-o-meter forum discussion:
@bubblesaurusrex—congratz! Until now there’s only one person who listens to music which is stranger than yours: joanofarctan