London calling: Musing about crowdsourcing

London calling to the underworld
Come out of the cupboard, you boys and girls

The Clash, London Calling

I’d heard about an unusual little burst of activity on Wikipedia over the last few days: people were frenetically editing and improving an article listing songs that were about London or parts of London.

You can find the article in Wikipedia here.

I found the rules for inclusion and exclusion fascinating.  I quote:

This is a list of songs about London. Instrumental pieces are tagged with an uppercase “[I]“, or a lowercase “[i]” for quasi-instrumental including non-lyrics voice samples.

Included are:

• Songs titled after London, or a location or feature of the city.
• Songs whose lyrics are set in London.

Excluded are:

Songs where London is simply name-checked along with various other cities (such as “New York, London, Paris, Munich”, lyrics of “Pop Muzik” by M

It struck me that this was an excellent example of crowdsourcing, of applying opensource processes to a research task. More importantly, based on the qualifying criteria, it seemed to me that here was a case where the crowd will always beat the computer, where it was actually quite difficult to write a program to compile the list.

Comments? Any other examples?