Ellyssa Kroski gets the scoop on social media, cloud storage, and Semantic Web initiatives at the world's largest library

The Library of Congress (LOC) has established itself as one of the leading institutions making use of social media to engage audiences and build community. And they have made great strides since January 2008 when they first began their Flickr Commons pilot. They have since launched a blog, a Facebook Page, YouTube and iTunes channels, and a Twitter account which claims over 13,000 followers. Most recently they have begun experimenting with the Semantic Web/Linked Data and cloud computing. I was fortunate enough to have a chance to chat with three of their librarians about these initiatives. The first set of questions were posed to Michelle Springer, a project manager of digital initiatives within the LOC’s Office of Strategic Initiatives. She oversees the Library’s Flickr pilot project, and has responsibility for the development and inter/intra-agency coordination of institutional policies related to social media pilots. Q.) Nearly everyone has heard about your Flickr Commons project at this point, could you tell us where you are with that and how the project has developed since its inception? A.) Our historical photographs continue to thrive in Flickr. To learn about the project in depth, please check out our October 2008 report, “For the Common Good: The Library of Congress Flickr Pilot Project”. We continue to add new images each week to keep the content fresh and interesting, and receive about 800,000 views a month for the 6,700+ images. In June we began offering something a bit different—a set of cover pages from the New York Tribune illustrated Sunday supplements, starting with the year 1909. The pages are derived from the Chronicling America newspaper resource  and complement the photos in our News from the 1910 set. One of the most interesting developments reflects the essence of Web 2.0 communities. A very engaged community-driven Flickr discussion group emerged spontaneously to focus on the Commons. They facilitate cross-fertilization of images between Commons collections and regular Flickr accounts and also write frequently about activities at participating Commons institutions on their blog. They even develop new software tools for Commons collections. Q.) In addition to starting a blog, Library of Congress has recently established channels on YouTube, iTunes, Facebook, and Twitter, what has been the response to these initiatives? Have you experienced any significant ROI on one platform vs. another? A.) The reception for the Library’s collections and news about Library events, acquisitions, and exhibits has been consistently welcoming and positive in all these communities. The Library’s Twitter account has over 13,000 followers; the Library’s Facebook page gained 5,000 fans in under a month; more than 24,500 Flickr members have designated the Library as a “contact.” Beyond these metrics we are interested in other measures such as the level of engagement and how grassroots enthusiasts become ambassadors for our content. Are our followers re-sending our tweets to their personal networks or embedding our images/video content on their Web pages or blogs? There is strong evidence that they are. For example, a Library video recounting the history of Rosie the Riveter that has been available for more than five years on loc.gov had received about 20,000 total views; that same video had received approximately 13,500 views over five months on YouTube, largely the result embedding across the Web. Q.) How do these recent social media initiatives fit into the Library’s overall strategic vision for the organization? A.) I’m going to quote Matt Raymond, the Library’s Director of Communications, and the author of the Library of Congress blog and the Library’s Twitter posts:

Social media represent a revolution in the way people interact and consume information. An entire generation is growing up with an expectation that digital content be made easily accessible and useful to their individual needs—outcomes that directly relate to the Library's historic mission. As the information marketplace atomizes and diversifies, and as public demand for transparency and responsiveness in government increases, it's incumbent on any organization or institution to adapt to the tools that are being used by their customers and audiences to communicate.

Establishing a Library of Congress presence in these social media communities helps meet various outreach-related strategic goals: they create conduits for individuals that might not visit loc.gov to experience Library collections and they increase opportunities for those individuals to interact with that content (by rating, commenting, and/or tagging it), and share it with others in their online communities. Q.) Can you tell us a bit about what goes on behind the scenes with regard to the process of choosing and getting content prepared for sharing on these social media sites? A.) There are two important guidelines we follow in all our social media pilots; we only post content for which there are no known publication restrictions and it must available on the Library’s Web site. We have also kept resource investment minimal, choosing content that is already digitized. Within those strictures, content has been chosen for the pilots that represent both the variety of materials we have in the collections and the events we host at the Library. Selecting and preparing this content are collaborative efforts and involve small teams of staff from across the institution including stewards of the collections, technical specialists, communications and Web services staff, and sometimes Library counsel. Requirements for content preparation have varied by media and the nature of the project. For example, many of the videos we've been delivering via our own Web site are in formats not compatible with the social media sites. Preparing this material often involves going to tapes or digital archives and re-encoding the videos. It is also important to consider what, if any, supplementary information might be necessary when content from the collections are experienced outside of the context of the Library’s Web site. For example, the Voices from the Days of Slavery: Stories, Songs and Memories podcasts (available at http://www.loc.gov/podcasts/slavenarratives/index.html and on iTunes U) are not solely historical recording, they include introductory remarks by American Folklife Center experts providing background and context for the recordings that follow. Additionally, as this content is shared or embedded, we want ensure that its provenance is apparent and so, in the case of videos, we include the Library’s brandmark at the start and finish. Q.) It looks like you are using your Twitter feed to automatically tweet your blog posts and drive traffic back to your blog—what other social media marketing tactics have you found useful? A.) You’re correct; we do try to leverage the value of our social media content using multiple distribution channels where it makes sense to do so. But original posts to Twitter outweigh the number of RSS feed postings. If all we were doing was redistributing the same communications, I think these sites would not be as popular as they are. We encourage subscriptions and provide incentives for repeat visits to Library accounts by continuing to load new interesting content, while developing workflows that will allow us to do this with minimum resource investment. And throughout all channels we provide links back to the Library’s Web site where more detailed information about collections, and sometimes higher resolution images, can be found. Another strategy is to align our social media presence with important Library events or exhibits when feasible. For example, we launched a very well received set of treasured portraits of Abraham Lincoln on our Flickr page  to coincide with the bicentennial of Lincoln’s birth and highlight the opening of the Library’s Lincoln Bicentennial exhibition, “With Malice Toward None”. And the Library’s blog  offered new images of the Library’s Lincoln bible when interest was at a peak (it was selected for use by President Obama at his inauguration), and now that we have developed multiple channels, we look for logical overlap opportunities, like adding a Flickr badge of our content to our Facebook page. Q.) Library of Congress has really jumped into the Web of social media with both feet, were all of these initiatives part of a pre-planned rollout? A.) No, although they all fall under a strategy to take advantage of tools available to increase exposure and sharing of the content available in Library collections. For each of these pilots we established business goals then determined the distribution channel(s) through which to initially experiment. For pilots employing external channels, the Library’s legal counsel had to negotiate changes to their terms of service agreements to modify provisions that were problematic for federal agencies, and completion of those agreements really dictated any rollout schedule. We are able to leverage work and experiences from one pilot to the next, however. For example, the comment moderation policy developed for the Library’s blog informs our moderation of community content added to our Flickr images. Likewise, monitoring and responding to the Flickr community-generated content will informs decisions about resource demands likely to be incurred if we undertake a similar crowdsourcing projects in other venues. Q.) What can we expect next from Library of Congress? A.) Here’s a quick list: Expect to see more collection-focused Library of Congress blogs in the next few months. The Library will soon be offering mobile phone users the option to receive SMS text alerts about the National Book Festival, and that will be a first for the Library. We’re exploring ideas about how we can engage those 5000+ fans on the Library’s Facebook page to encourage reading; perhaps through online book discussions aligned with Center for the Book events. And we’ll be launching a new literacy promotion Web site in September: Read.gov. One of its features will be the exclusive publication of new a serial adventure story written by a chain of well known authors, many of which will appear at the National Book Festival.

---

The next set of questions I asked of Sally McCallum, chief of the Network Development Office at LOC. Her Office is responsible for development and management of technical standards in collaboration with the national and international standards bodies and with the information community. The Office carries out research and development into new technologies for using and presenting data, especially related to the digital environment. Q.) Very few people have heard about the Library of Congress Simple Knowledge Organization System (SKOS) project, could you give us a brief description of what it entails? A.) The Library of Congress Authorities and Vocabularies service, our “SKOS project,” enables machines—and humans—to programmatically access authority data at the Library of Congress. This service is related to the “Linked Data” movement's approach of exposing and inter-connecting data on the Web via dereferenceable URIs. Users (machines) request data via URI in this service, and then are able to specify the serialization of the data that is returned to them. The choices include the RDF serialization Simple Knowledge Organization System (SKOS). While there are or will be other serializations made available (e.g., JSON, MARCXML and MADS), the first vocabulary in this service, the Library of Congress Subject Headings (LCSH), is SKOS compatible so SKOS is a primary option. At present some details about the LCSH concepts that are present in the MARC version of the data that are lost in the SKOS version. LC will be working on extensions to enable full values, while staying within the SKOS framework. The eventual scope of the Authorities and Vocabularies service is to provide access to commonly found standards and vocabularies promulgated by the Library of Congress, such as the, Thesaurus of Graphic Materials, MARC Geographic Area, Language, and Relator codes, and several code lists from key standards such as the Preservation Events and Preservation Roles from PREMIS or various MODS and MARC code or term lists. Q.) Why do you think this type of initiative is important at this time? A.) Put simply, SKOS is an approach and special set of data tags that are being developed by the Semantic Web community with the hope and expectation that they will help enable interesting semantic manipulation of data via machines in the future. The primary audience is the machine, but most implementations of SKOS will provide a Web human interface also as a byproduct, as LC did. For the machine user of the service, the SKOS version of the data will enable linking of the vocabulary terms in local records with the LC authority record via Linked Data. This linkage may be utilized in a number of ways including automatic update of values “behind the scenes” by the machine. The possibilities enabled by SKOS and related semantic tools need to be explored. Not all will be viable, but some will provide new pathways forward. The provision of the Library of Congress vocabularies in SKOS, in addition to well known library exchange formats, can facilitate the use of our data by the broader internet-based community. We cannot predict the future but we can support exploration and experimentation with important standards by both the library and non-library communities.

---

The final two questions were posed to Bill LeFurgy, the Digital Initiatives Project Manager with the Library of Congress. He is responsible for working with libraries and archives around the country in connection with the National Digital Information Infrastructure and Preservation Program. He also represents the Library in digital preservation collaborations with other nations. Q.) You have recently partnered with DuraSpace for a one-year pilot program testing out cloud storage capabilities, can you tell us a bit more about the initiative? A.) The Library believes that permanent access to digital content demands a variety of tools and services. We want to see how cloud computing meshes with this approach. Our partners will test DuraSpace to learn how effective it is for storing, searching, and retrieving different types of data. We are especially interested in studying costs to see if cloud computing is economical for large-scale digital stewardship. Q.) What type of an impact do you think cloud computing could have on the library field? A.) It has great potential. Cloud computing offers the prospect of a distributed preservation infrastructure, which is vital because no single institution can handle the job alone. Most cultural heritage organizations lack the resources and the capabilities to preserve large amounts of digital content—and the volume of data worthy of preservation is growing by the day. Libraries will need to work together as part of a collaborative network to achieve the necessary economy of scale. Services like cloud computing fit nicely within this collaborative concept. The Library of Congress continues to blaze new trails and it seems that we can expect even more exciting developments to come. I want to thank Michelle, Sally, and Bill for taking the time to share their insights with everyone. Ellyssa Kroski is an information consultant, reference librarian, writer, and conference speaker, as well as an adjunct faculty member at Long Island University, Pratt Institute, and San Jose State University. She blogs at iLibrarian.