The Flickr Commons: Collaboration and Crowd Sourcing

The Flickr Commons was started in 2008 as an effort to make the publicly held photograph collections held by various institutions around the world more accessible to the public. The first collections were provided by the Library of Congress, who released almost 3000 images when the project opened. They were soon joined by other institutions, and now The Commons boasts over a hundred participating institutions, each with thousands of photographs of their own.

The Commons is a fascinating example of both a large-scale collaboration between institutions internationally and a crowd sourcing effort. Each institution brings its own collections to the table, but they are all accessible through Flickr with the same purposes and goals in mind, to make the photographs accessible and to bring the public into the conversation, so that they could bring information to the table that the institutions themselves couldn’t provide. The process allows the public to become involved in the collections, and exposes them to users and makes them accessible to people who otherwise may not have known about their existence at all.

The process is simple to get involved in, though it does require some knowledge of Flickr and how it works. It’s free to register an account, but you don’t need one to access content. To browse the photographs, you have to go to an individual institution’s account and look through either their photostream or albums, though it’s also possible to search Flickr using the tags. Because tags are provided by the user base and there’s no controlled vocabulary, it can make searching for specific photos difficult, but Flickr embraces the browsing mindset, and institutions generally try to follow that example. The Flickr Commons isn’t meant to be a comprehensive database, but rather a method of sharing and interacting with users.

As with any crowd sourcing effort, there are hesitations and concerns, ranging from worries about a lack of a controlled vocabulary making the crowd sourced metadata inconsistent to a lack of policing in some cases. For the most part the Flicker Commons project has been widely accepted and lauded, but there will always be concerns.

What I think a lot of the detractors of the Flickr Commons crowd sourcing are missing is that the institutions who participate in the Commons aren’t relying solely on the information contributed by users to form the metadata for their collections. Tags and comments and notes don’t immediately go straight to the item’s original institutional data. Flickr Commons is a method for institutions to reach out to users and incorporate them into the process, and to make their collections accessible to everyone freely. There has been a lot of incredibly useful and meaningful additions provided by Flickr users, but ultimately it all has to go through employees of the institution, who generally research each addition before editing the metadata of the original object. So when a photo gets drowned in inane notes and cryptic or meaningless tags, it’s unfortunate, but doesn’t affect the original object at all. The two main goals of Flickr Commons, as stated on the Commons’ main page, is “1. To increase access to publicly held photography collections and 2. To provide a way for the general public to contribute information and knowledge. (Then watch what happens when they do!)” Naysayers, such as Larry Cebula in his blog post “Lick This: LOC, Flickr, and the Limits of Crowd Sourcing” complain about how images can get buried in ultimately unhelpful tags, notes, and comments, but they don’t seem to be willing to look beyond the front page of the Commons. The Flickr Commons community responded to this post and discussed it, and many users pointed out that once you get past the featured items the tags and comments become much more helpful. It’s important to remember that in this sense, Flickr is still very much a social media website. The community as a whole uses it that way, and whenever anything is opened up to the public like this, you have to be prepared to wade through hundreds of “great pic!” comments to get to the users who are identifying people and places. Dedicated users, who are interested in contributing to the information effort, will spend more time browsing deeper into the collections and provide tags and comments, which will then allow other interested users to discover the photos. Casual users will see photos on front pages and react, then move on. It’s just how social media works, and institutions and professionals who are observing effects need to be prepared for it.

            It is interesting to see that for the most part, LIS professionals are praising the effects and reaction to the Flickr Commons. While there are some detractors, like Cebula, most commentary on the project is positive. The authors of “Rethinking Evaluation Metrics in Light of Flickr Commons,” all employees of various Commons institutions who are remarking on metrics and effects, discuss how interaction between institutions and Flickr users enables for a deeper sense of participation, and allows users to become more invested in the outcome. Users work together to unravel mysteries of who what when where and why, and institutions respond and update their records accordingly. One of my favorite examples comes from the New York Public Library, where one user asks what was wrong with the arm of a soldier in a photograph, and another user responds and answers the question by linking to another photograph in the collection which provides a diagnosis. This community involvement is a big reason why Flickr was chosen in the first place, since at the time LC was considering where to host the project it was the one photo sharing site with the most of that kind of community involvement.

            The Commons is still very much alive and strong, though the reported metrics I found were from 2011 at the most recent. I expect that’s because professional curiosity has mostly died down since its inception. Regardless, one of the interesting problems that institutions report (again in the “Rethinking Evaluation Metrics…” article) is that it’s difficult to fully gauge metrics on the Flickr Commons. They can track views, followers, number of comments and added tags, but measuring “significant engagement” is qualitative and difficult to measure. Each institution remarks that they have received plenty of engaged users, but they have trouble tracking particularly significant tags and comments, or references to an image outside of Flickr. In general, institutions agree that while it’s difficult to measure qualitative experiences, the project is well worth the costs and they’ll continue to operate.

            I found some excellent resources while investigating this topic, which I’ll include now. "Rethinking Evaluation Metrics in Light of Flickr Commons" was a great article written by staff from five different institutions involved in the Flickr Commons: the Powerhouse Museum, the Library of Congress, the Smithsonian Institution Archives, Cornell University Library, and the New York Public Library. Each institution offers a different perspective on metrics they use to gauge user involvement and interest, and shares stories of user involvement that are fascinating and entertaining.

           Newsweek wrote a short article on the initial effects of the Flickr Commons project, focusing on the Library of Congress. It's a good introduction to the project from the LC's standpoint.
          Evan Fay Earle's thesis "Crowdsourcing Metadata for Library and Museum Collections Using a Taxonomy of Flickr User Behavior" provides a good look at some of the benefits and challenges of using the community of Flickr as a crowd source.
         Larry Cebula's blog post "Lick This: LOC, Flickr, and the Limits of Crowd Sourcing" is a detractor's standpoint, though the argument is somewhat narrow. What I found more interesting and helpful was the Flickr Commons community discussion in response to it, where they intelligently commented on the drawbacks of using a community as large as Flickr to crowd source metadata, and generally concur that while there are difficulties and hurdles, the ultimate benefits outweigh the concerns.
       Finally, the article "Smithsonian Team Flickr: a Library, Archives, and Museums Collaboration in Web 2.0 Space," by Kalfatovic, Kapsalis, Spiess, Van Camp, and Edson, available on LIS Source through the Simmons Library, was an insightful article on the collaboration across the many facets of the Smithsonian Institution to share separate collections together in one Flickr Commons space.