Using the Crowd
I recently watched a really great video of a presentation given at the National Archives about several archival crowdsourcing projects. Crowdsourcing utilizes the groups of people to accomplish a task, which is often information related. When you ask your contacts a question on Twitter or Facebook, that’s a small scale version of crowdsourcing.
The presentation (which is an hour long, but definitely worth a watch if you’re interested) included looks at three projects using crowdsourcing that are currently being run by archival institutions.
- The World Archives Project, Ancestry.com: indexing vital records to make them more searchable for genealogists.
- The Map Warper Project, New York Public Library: aligning and layering historical maps on modern maps.
- The North American Bird Phenology Program, US Geological Service: capturing historic bird migration data
All of these programs are very large in scale and address the limitations of technology when working with archival materials. There are just some things that a computer can’t do as well as people! In particular, optical-character recognition software cannot handle handwriting very well and without metadata, a computer can’t do much with an image. Another crowdsourcing project that has been making a lot of waves lately is the New York Public Library’s menu transcription project, called “What’s on the Menu?”. Yet another, Flickr Commons, invites anyone to comment and tag images from archival collections.
Archives have used volunteers for a long time to help – so what makes crowdsourcing projects different? These projects can also be viewed as micro-volunteering opportunities. Since they are accessible from home, volunteers can contribute to the projects at odd hours. Typically, the work is broken into small chunks, so each individual contribution has a small time cost to the volunteer. This allows volunteering to be possible for those that are otherwise extremely busy. Scale is another factor; these projects can incorporate many, many volunteers simultaneously and address the problems of very large record groups. For the larger of these projects, both the volume of records and volunteers number in the thousands.
On the other hand, what impact will these types of projects have on small, local institutions? The records and technology behind these projects is large in scope, so only the bigger institutions will be able to attempt such an undertaking. Will they draw potential volunteers from assisting their local archives in person, or will they encourage volunteering in the community by sparking an interest in historical records?
Furthermore, is the effort involved in setting up and monitoring the systems involved in producing these projects more efficient than other strategies? I think they can be; crowdsourcing projects seem to be gaining in popularity, and the archival community is learning a lot about how to make them successful. But, institutions also need to use appropriately trained staff to coordinate and care for these projects.
What’s your take on crowdsourcing? Have you participated in a project like this? Do you know of any other great crowdsourcing projects? Tell us in the comments!