Crowdsourcing Games: 1 June 2015

In the summer of 2013, photographer James Dee donated his collection of 150,000+ images to Artstor. These images, taken over his 40-year long career, documented New York City’s contemporary art scene and were destined for the dumpster. No one was willing to accept and archive these images for one simple reason: there was no metadata.

James Dee’s problem is not unique. Many of the challenges faced by archivists, catalogers, and curators are caused by metadata (or lack thereof) and the limited availability of human and time resources to supply and correct this information. As a small non-profit, we also faced this challenge and decided to engage our most valuable resource: our community of users.

Enter Arcades

We know that most people (even enthusiasts) won’t want to spend their free time tediously entering metadata. The solution? Make it a game!

Arcades (Walter Benjamin anyone?) is our take on a metadata crowdsourcing application. Many institutions such as the British Library and the Boston Public Library have explored this idea and indeed, the concept is not new. The success of Wikipedia or OpenStreetMap raises the possibility of harnessing collective knowledge.

Our approach with Arcades is to simply expose the central act of entering metadata, and frame it within a system that would track progress. This differs from other approaches where the act of entering metadata is obscured or wrapped within a more traditional “game” experience or aesthetic. We believe that a simple, repetitive action can be the foundation for an addictive experience. These experiences can be found in games of enduring popularity such as Pong, Tetris, or Angry Birds or even in physical experiences such as eating candy bars, chips, or sushi.

The design for Arcades is based on a piece of paper, turned on its side. As images flow onto the page, the user (the gamer) enters metadata. The entire game can be played without the user’s fingers leaving the keyboard to provide for a fast and fluid experience.

Gamers collect points based on the number of fields and types of fields they fill out. As we learn more about the types of metadata we want to collect, we can adjust these points to reflect difficulty/usefulness/correctness of the data. Total points are aggregated/displayed and allow a user to level-up from a “Dilettante” to a “Master.” Arcades also displays all images that a gamer has successfully cataloged. These are saved from session to session to allow a user to build up their unique “collection.”

During the first phase of the project, our goal is to collect as much metadata as possible. Our analysis of this data will inform our adjustments to the scoring system and gaming experience and allow us to build additional methods for guiding users to provide appropriate responses.

We encourage you to participate as a beta tester! Sign up now at: