Wednesday, July 11, 2007

The reason for Last.fm's popularity

I've been thinking about last.fm and about how such a fairly simple concept has so many people, including me, going "wow that's so cool!".

A lot of the stuff at last.fm is pretty standard web 2.0 stuff: friendslists, blogging, tagging, etc, but the key feature that sets it apart in my eyes is the scrobbling: the way it aggregates the music you listen to and provides handy charts to display based on this aggregation.

The coding to collate and display this data probably isn't that complicated, but the result seems to me to have a social resonance that goes well beyond something I'd expect from a "mere" aggregation of data.

Part of it's obviously the type of data. Musical preference is a much more personal thing than, say, a person's preferred shopping habits, so that has an impact. But I think it's also the way in which we talk about music that allows last.fm to augment music buff's intercommunication about music so effectively.

There are two main questions that a music buff wants to answer, even without being asked: "what are you into?", and "what are you listening to right now". Last.fm's overall chart and weekly chart provide a neat answer to both. The assumption is that your musical preference can be quantitively scaled by collating the number of times you listen to each track and each artist. The process can be cheated obviously, but the overall result translates extremely well: Artists and tracks you like score highly while less well-liked artists score less well.

It seems so obvious, but I only really thought how useful it is to do that when I saw Last.fm (then Audioscrobbler.com) doing it. A few other things that make the process so effective:
It's transparent. Once the plug-in for music player X is installed, the process of collation(scrobbling) is automatic. Many of the modern linux players even have the process built-in, so that commencing scrobbling is as simple as checking a box then forgetting about it.
It's efficient: the time it takes to detail musical taste, even to the point where you can point out what's currently tickling your eardrums and providing an approximate ranking of your taste, is practically instantaneous.
It's emergent: neat word that. Basically it means that you don't know what the chart's going to look like until after it's completed. You can actually be surprised by the results and think things like "wow, I didn't know I was listening to Muse that much!".
It's "push": I guess by this I mean that you can provide this neatly aggregated and quantified snapshot of your musical taste to people without them asking for it. Just lay it out on your blog, Myspace or whatever and you answer the two questions above to anyone who looks at your personal space whether they were going to ask you or not.

I guess overall that it seems to me like last.fm combined some fairly obvious ways to use software in a manner that allowed people to communicate about music in a way that had never been done before. That really is cool.