Is Wikipedia Loves Art Getting "Better"?

It's rare that a participatory museum project is more than a one-shot affair. But next month, Britain Loves Wikipedia will commence--the third instance of a strange and fascinating collaborative project between museums and the Wikipedia community (Wikimedians). The project's implementation keeps changing, and I can't decide whether it is getting better or just different. I hope you'll share your thoughts in the comments.

Wikipedia Loves Art, Take One

The first version of Wikipedia Loves Art first took place in February 2009. It started with a request from a group of New York Wikimedians to the Brooklyn Museum. The Wikimedians asked if the museum would coordinate a project in which people could photograph artworks in cultural institutions to illustrate Wikipedia articles. The museum agreed and brought fifteen institutions from the US and UK on board to participate.

The museums asked to Wikimedians to provide the institutions with lists of thematic topics that required illustration. Museums used these thematic lists to develop scavenger hunt lists to distribute to participants so that they might find art objects to illustrate Wikipedia topics like "Roman architecture" or "mask." Participants were asked to photograph objects and their accession numbers so staff could identify and describe the objects properly. The museums developed careful rules about what could and couldn't be shot, and how participants could upload their images to Flickr for use by the project.

The event succeeded in donating thousands of images to Wikipedia, but it was plagued by challenges that frustrated museum staff, Wikimedians, and photographers.

Some of these challenges were about mission fit. The Wikimedians’ and museums’ goals were not as aligned as they originally thought. Museums saw this project as an opportunity to engage local photographers to think creatively about how artworks might represent different topics. In contrast, the Wikimedians were focused on making cultural content digitally available online using as open a licensing structure as possible. The museums cared about participants connecting with artworks and identifying them properly, whereas the Wikimedians cared more about participants sharing images under open legal licenses.

Other challenges derived from the complicated and shifting setup (see, for example, this discussion about clarifications to the rules). From the institutional perspective, the best way to deliver good participant experiences was to constrain contributions through the Flickr uploading system. Staff were concerned about losing control of images of their collections, and they wanted to make sure the images were linked to the correct information about each object. But there were many Wikimedians who were confused or frustrated by what they perceived as arbitrary institutional constraints in the submission format. Some people found their own rogue ways to upload museum images outside of the project framework, much to the consternation of museum representatives, who saw these actions as causing confusion and potentially violating intellectual property agreements.

The project's setup also created more work for participants and museum staff than anticipated. Over 13,000 photographs were submitted by 102 photographers at the fifteen different institutions, documenting about 6,200 pieces of art. While these participants had done the hard work of capturing the images, it was up to the institutions to validate, tag, caption, and prepare them for Wikipedia's use. Image verification was a Herculean effort. Using accession numbers to identify the objects was not successful; it was complicated for participants, and staff were not able to verify many images. Some images were disqualified for copyright reasons, others because they could not be identified. Eventually, all of the work was completed and 6,195 photographs were donated to Wikipedia. But when the dust settled, the overall effort for institutions involved in Wikipedia Loves Art was so great that many saw it as an unsustainable collaboration.


Wikipedia Loves Art, Take Two

In June of 2009, Dutch Wikimedians tried again. They partnered with 46 institutions in the Netherlands to produce Wikipedia Loves Art / NL, which took a different approach to the project. Rather than starting with a list of themes provided by Wikimedians and inviting visitors to shoot the objects that they felt fit the topics, the Dutch Wikimedians asked the museums to provide a list of specific objects that participants could photograph. This compromise achieved three things:

1. The museums knew exactly what would be photographed and could more tightly control the experience. At some institutions, staff members set up specific dates for photography and escorted photographers through the galleries.
2. The Wikimedians knew that all of the images would be legal for use from a copyright perspective. There was no concern about museums needing to verify that an object on the list was in fact legal for use.
3. The participants received a numbered list of objects to photograph and could tag their images with these id numbers instead of with accession numbers. This significantly reduced the number of object identification errors and reduced the staff time required to review the images submitted.

Wikipedia Loves Art / NL also included some other changes, most notably a centralized website that coordinated all of the events and information. It ended June 30 with 292 participants contributing 5,447 photographs. There were still validation errors; for example, the "winning" photograph from the project was discovered to have been taken at least a year prior to the event. But in general, the project went smoothly. The images were uploaded more quickly with fewer staff hours than in the original version, and the institutions and Wikimedians considered it a success.


So is it better?

From my outside perspective, Wikipedia Loves Art / NL more successfully served the needs of the museums and the Wikimedians than the original event. But it was also fundamentally different for participating photographers. It offered participating photographers less creative agency and less responsibility... and less attendant confusion.

Did Wikipedia Loves Art get better? The answer to that question depends on your values. It got better at meeting the partners' needs, but worse at allowing individual participants to determine the outcome. I also wonder whether the changes impacted the extent to which participants felt connected to the institutions as opposed to seeing the museums as venues for a project. Ann Beaulieu, a researcher associated with the Tropenmuseum reflected on discussion in the Wikipedia Loves Art / NL Flickr groups, commenting:

the goal is to get photos of the museums’ collections onto Wikipedia. Interestingly, this does not seem to be obvious to some photographers who see photo-making as the ‘end’ or goal of their practice, and consider getting objects in museums photographed and into Wikipedia as secondary.

To me, the newer version of Wikipedia Loves Art seems less suited as an onsite audience engagement program for photographers, but more suited to provide Wikimedia with useful data. This may make it a better project (more useful) or a worse project (less engaging for participants).

I realize, however, that this perspective reflects my own value judgment that local community engagement is a more important part of museum missions than providing digital access to content. There is a large community of people who will use the digital images in their new home on Wikimedia Commons, and who am I to say they are not equally important?

I also am biased in thinking that giving participants creative agency is "better" than giving them a list of photographs to shoot. You could just as easily argue that the project got better by making participation simpler, and that the redesign DID serve the participants who are part of the Wikimedia community.

What do you think?