Jen Wolfe 2015-01-26 17:58:40
The potential benefits of crowdsourcing for cultural heritage institutions are legion. Along with the expected enhanced data and increased public engagement, bonus results may include ongoing media attention, an uptick in new gifts, and stronger relationships with existing donors. With its DIY History site (http://diyhistory.lib.uiowa.edu) that invites the public to transcribe manuscript letters, diaries, and cookbooks to enable full-text searching, the University of Iowa Libraries (UIL) has been lucky enough to experience all of these crowdsourcing benefits and more. UIL started its first crowdsourcing effort as an experiment in spring 2011, uploading a few dozen soldiers’ diaries for transcription as part of a Civil War sesquicentennial project. Nearly four years and more than fifty thousand pages later, transcription crowdsourcing has become a major strategic initiative— informing operations from acquisition and digitization priorities to equipment purchases and even hiring decisions. As a librarian in UIL’s Digital Research and Publishing department until very recently, I spent a good chunk of my career over the past few years helping to develop DIY History. Here’s what I learned. January 6, 2011: Dear Diary, Today Greg asked about creating a transcription crowdsourcing site to promote our Civil War digital collection. I said I’d investigate (AKA Google it, since I had no idea what he was talking about). January 11: Sounds like the project would take way more programmer time than we’re able to commandeer, alas. January 13: Library webmaster and girl genius Linda to the rescue! Her workaround: a web form, an email submission box, and lots of copying and pasting. Low-tech and high-touch, but functional! May 5: Launched! I wonder if anyone will use the site? How sad would it be to throw a crowdsourcing party where nobody showed up? June 1: My mom likes it....June 7: Success! Featured on the American Historical Association blog, site usage ensued. June 8: Too much success! Today we’re on the front page of Reddit. Goodbye, servers.... Year One Lessons Don’t be afraid to experiment. Libraries and archives are risk-averse environments; it’s become cliché to say that our professions need to become more entrepreneurial, but that’s because it’s true. If you don’t think you have enough resources for an experimental project, try it anyway—reallocate, reconfigure, make it work. Choose content wisely. As scholar Adam Kriesberg cautions: “Part of what makes a good transcription project is not the transcription in and of itself but the existence of a compelling, historically interesting collection that needs crowdsourced labor to be fully utilized. Without that, or in cases where we just throw up any old collection online for users to poke at, we are missing an opportunity.”1 Promote ceaselessly. After our first press release was universally ignored, we targeted Civil War listservs and historical associations until eventually finding our audience. Once you’ve found yours, keep going! Use traditional and social media to keep projects in the public eye. Extreme weather? Black History Month? National Pie Day? Search the transcriptions for relevant text and share. Recent acquisitions? Project milestone or anniversary? Launching a new collection? Fire up a press release, blog entry, Tweet, Tumblr post, what-have-you, and let the world know. June 12, 2012: Civil War papers are complete, so we’re redesigning the site as DIY History, with multiple collections, an actual infrastructure (the open-source Omeka content management system and its new transcription plug-in, Scripto), historic photo tagging, and more! August 2: Behind deadline on the software implementation, so for the next month we have all of Matthew’s and Shawn’s time, i.e. two full-time developers, just like a real, grown-up DH center. October 9: Relaunched! Promoting the new site around our manuscript cookbooks, a departure from previous narrative-based documents. I hope the crowd still follows? November 23: Success! Lots of internet coverage, including CNN.com, but once again a social media link-sharing site—this time MetaFilter—was the big break. December 15: What if you threw a phototagging party and nobody showed up? Year Two Lessons Use early success to make a case for more resources. Once we demonstrated a demand for crowdsourcing, it became much easier to access scarce library resources, such as programmer time. Even small-scale pilots or proof-of-concept designs can go far in gaining project support. Try out new approaches and see what sticks. Inspired by initiatives like the Library of Congress–led Flickr Commons, we added tagging functionality for historic photo collections to our site, but eventually removed it after little uptake. However another departure—the inclusion of handwritten cookbooks in addition to letters and diaries—was well received, inspiring future plans for transcription of science fiction fanzines and natural history specimen cards. Pay it forward. We happily abandoned our labor-intensive submission process, but the cost was a different kind of labor—customizations to Omeka and Scripto to make the activity of transcribing more prominent, simplify the user interface, and add progress bars for contributors to see the impact of their work. It turns out we weren’t the only ones interested in these modifications; after our programmers made the revisions available on the code-sharing site GitHub, they’ve been used by a number of other institutions. February 17, 2013: Brainstormed with Kelly from reference and Matt and Tom from rhetoric about a month-long curriculum module using DIY History. The plan: a scalable and open-access assignment for potential use by any of our rhetoric instructors and their (gulp) four thousand–plus first-year students—or for adaptation by instructors from other disciplines or institutions. September 9: After intense Planning this summer, the project has a name (Archives Alive!) And a few sections of Honors Rhetoric to test drive it. September 24: Students transcribing correspondence donated by Evelyn Birkby came up with an unexpected source for historical context—Evelyn herself! They interviewed the ninety-four-year-old author, journalist, and radio personality via email and conference calls. She told our curator Kären that she loved participating in the project, because young people make her feel “peppy.” Year Three Lessons Seek opportunities to collaborate. While gaining audiences around the world, we had no luck on campus until finding likeminded partners interested in classroom innovation through digital pedagogy. By combining our areas of expertise, we helped each other while advancing university strategic goals around undergraduate engagement and developing digital humanities initiatives.2 Don’t forget donors. Along with attracting new donations of family papers, the crowdsourcing project has strengthened existing relationships—engaging those who get to see their papers online and in use. Reach out to researchers. Pre-crowdsourcing, you wouldn’t want to send casual users digging through box after box of probably irrelevant manuscripts. But thanks to the magic of full-text search, they can be directed to the transcriptions for an instant survey of public opinion on any number of historic topics. March 17, 2014: Scanning and more scanning of war letters and diaries for this summer’s WWI centennial. July 29: Oops, missed the anniversary, not enough content online yet. Mass digitization is extra challenging in the summer with staff traveling for conferences and student workers scattered to the winds. September 20: After this semester’s three hundred–plus freshmen finish up Archives Alive!, we’ll launch the collection-in-progress— three thousand pages so far, fewer than we’d like, but it should hold us for a while. October 5: Not holding! BuzzFeed then NBCNews.com featured DIY History, and the resulting spike of visitors pretty much transcribed the entire site in a few days. Preservation has all hands on deck in an attempt to keep up with demand, but it’s not a fair fight. Year Four Lessons Expect the unexpected. Post-BuzzFeed, our IT staff asked about patterns or predictability for the traffic spikes that stress library servers in addition to scanning staff, but it doesn’t exist. Sometimes there’s overloading and crashing, sometimes it’s crickets, usually it’s somewhere in between. All we can do is try to be as nimble and responsive as possible when necessary. The future is already here, it’s just not very evenly distributed throughout the library. The lag between supply and demand for transcribable content is common to successful crowdsourcing projects, given the imbalance between relatively small-scale, staff-sourced digitization and scaled-up crowdsourced transcription. Continuing to modernize our workflows and further develop mass digitization methods should better enable us to keep pace with the crowd. It takes a village. Maintaining a transcription crowdsourcing project requiring coordination and hard work from staff in special collections, conservation, preservation, cataloging, information technology, and digital library departments isn’t easy. But the payoff—fulfilling our mission by actively engaging users and connecting them with their cultural heritage—makes it all worthwhile. January 2, 2015: Off to Chicago for my new job as Digital Initiatives Librarian at the Newberry Library. I’ll miss my UI colleagues like crazy, but I’m thrilled about all the new projects I’ll be working on, including one that already feels like home: the Newberry’s Civil War in Letters transcription crowdsourcing site (http://publications.newberry.org /civilwarletters), partly inspired by and reusing code from DIY History. Notes 1 Comment on “Crowdsourcing Cultural Heritage: The Objectives Are Upside Down,” March 10, 2012, Trevor Owens: User Centered Digital History <http:// www.trevorowens.org/2012/03/crowdsourcing-cultural-heritage-the-objectives-are-upside-down>. 2 See “Archives Alive!” open-source curriculum: http://ideal.uiowa.edu/projects/archives-alive.
Published by Society of American Archivists. View All Articles.