I recently happened upon a compelling crowd experiment called UrbanGems. The idea between UrbanGems is simple: a user views an a-b choice between two images and is asked “which is more beautiful?” The user is then asked what percentage of the population they think agrees with them on their choice and at the end of a ten picture series is assigned a points value based on how accurately they guessed agreement.
Right now the imagery is limited to London street scenes and nearby locations of a quality that is reminiscent of Google street views. They don’t look like polished photographic examples or curated selections – it all appears to be randomly generated, but whether that’s true or not, is not what is most interesting about the project to me.
What is interesting is the stated goal of the project. The team that has launched UrbanGems is trying to determine some subjective information: what are the most common visual cues among pictures that make one consider it beautiful. Can one truly measure something so personal? Certainly if they aim to have a large amount of sample data, they are off to a good start at illustrating that we may have some common patterns on what we consider beautiful (or also, quiet or happy). The team would like to correlate that to some social truths of opulence or deprivation and see how that might skew results.
The site just launched just over a month ago and it will be interesting to see how it evolves over time. Interested in sharing your opinion. Contribute to urbangems today. It might not be particularly complex, but the results, I hope, will be fascinating.