 Hello, I'm Marely Profitt from OCLC Research and I'm here today with Sarah Snyder from the Archives of American Art at the Smithsonian Institution and we're here today to talk about our experiences with Wikipedia, how libraries and archives are working with Wikipedia and to talk to you about our different projects. So the project that's been going forward at the Smithsonian Institution has a different flavor and nature than that which is going on within OCLC Research, but they do have some very important connections in common with one another. So I'm going to be talking about a little today about why Wikipedia is important and why this should be of interest to you. Well, the first thing is that it's really about the numbers. Wikipedia is the most highly ranked, one of the most highly ranked websites, number six in global traffic via Alexa, number five I believe in the United States. So a lot of people have their eyeballs on Wikipedia and we want for them to be able to find things in Wikipedia and then to flow into libraries. So making sure that library content, library data is represented in Wikipedia is very important just by the numbers. Another reason and perhaps just as important is that the Wikipedia community is really ideologically aligned with the library mission and that is that their tagline so to speak is really the same as ours which is to provide access to knowledge for free to our constituents. We want to be sure that we're providing access to information and making that as easy as possible and that's also the global mission of Wikipedia and the Wikipedia community. And the third thing and I think perhaps a little less known to librarians is how much Wikipedia's really share our appreciation of quality sources. So you may recall going to Wikipedia sites and you can see this little template at the top chiding the Wikipedia community. This article does not cite any references or sources. This ironically is an article on reference management software. Citing references and sources and citing reliable references and sources is incredibly important to Wikipedia's. So we too share that value of very reliable, incredible sources. So that's yet another thing that we have in common. But there are special challenges in collaborating with Wikipedia and building the types of collaborations that Sarah and I have participated in. First of all it can be really difficult to figure out who it is that you're collaborating with or negotiating with. The Wikimedia Foundation is not a place where Wikipedia's work. It's really there to kind of keep the doors open, keep the servers running, maybe identify some high priority projects for the Wikimedia projects overall. That's not who you approach if you want to collaborate with Wikipedia instead. You need to find collaborations with groups of some of the millions of editors within the community, certainly many thousands of active editors within Wikipedia. There's a problem there in that the community of editors is really distributed and virtual. Each Wikipedia editor is really motivated by their own personal reasons for participating and volunteering in this really unique and collaborative resource. So it's really quite a challenge to connect with individuals who are interested in collaborating with libraries and archives. But once you can find those people, real magic can happen. An additional challenge is this culture of combating what's known as link spam. So Wikipedia editors are very on guard for links to sources that may be really trying to promote their own search engine optimization by gaming Wikipedia in some way. So historically some cultural heritage organizations have had quite a challenge linking to resources, even resources that were highly relevant to the articles at hand because editors were overly perhaps on guard about linking. And perhaps there was some misunderstanding on the part of cultural heritage institutions as to what is appropriate linking within Wikipedia. So those are some of the challenges. An additional challenge is that library resources can be pretty hidden from a Wikipedia editor's point of view or somebody who works on Wikipedia. This is a site, WikiRank, which ranks the top sites that are linked to from Wikipedia. And you can see that Google Books and Amazon.com come in in the top 15. These are really easy sites for Wikipedia's who are looking on the web for authoritative incredible resources, easy for them to link to because they can find these sources. It is going against their own principles of giving access to free knowledge in Google Books. Of course, you will find open source content and that's terrific. In Amazon, everything that you find there is pretty much going to hit a paywall. I think that Wikipedians would prefer to link to libraries but our resources can be pretty invisible to them. So looking at this list way far down on the list, Worldcat comes in at number 81 in terms of sites that are linked to from Wikipedia and Hottie Trust is not even on the top 100. So we'd like to see what we can do to call more awareness to library resources within the Wikipedia community. So what can be done? Well fortunately, there is a group of motivated individuals within Wikipedia that refer to themselves as the Glam Wiki group and Glam is an acronym for galleries, libraries, archives, and museums. And when I was able to connect with this group, I found a number of kindred spirits who were really interested in having the collections, the contents of cultural heritage institutions such as libraries and archives, better represented within Wikipedia. So this is kind of our natural home and who doesn't want to be glamorous. So this is a good place to be. An outgrowth of the Glam Wiki project is the idea of a Wikipedian in residence. The first Wikipedian in residence was at the British Museum in 2010. It was a short term assignment, I think perhaps maybe six weeks. It was conceived of by a Wikipedian who thought that this would be a great idea if he could embed himself within a cultural heritage institution and he was very interested in having the content of the British Museum better exposed in Wikipedia. So he proposed this to the British Museum and fortunately they went for it. Since that time, there have been a number of Wikipedians in residence. The Smithsonian's Archives of American Art, Sarah will talk about that in just a minute. The Indianapolis Children's Museum, the National Archives in the U.S., National Library of Israel, British Library, and on and on and on. This is quite an international group not only representing a wide variety of cultural heritage institutions but also that international aspect which is quite nice. And the idea has really spread like wildfire. If you go to Wikipedia and look up Wikipedians in residence, you will see the range and number of residencies that have been developed and are in development even at the moment. So when I was looking for a way to work with Wikipedia, it struck me that having a Wikipedia in residence would really better inform our view of the opportunities for working with Wikipedia. Because what better way than to invite somebody who was quite skilled to who really knew the inner workings of Wikipedia to come work with us and tell us what they were interested in and not just work from a position of what we were interested in. So it's been a very fruitful partnership for us. So before I tell you about our story, I'm going to pause for a moment and let Sara tell you about her story. My name is Sara Snyder. I work at the Archives of American Art which is one tiny corner of the Smithsonian Institution, the world's largest museum and research complex. The Smithsonian includes 19 museums and galleries, the National Zoo, as well as nine separate research centers of which the Archives of American Art is one. So the Smithsonian is probably best known for its museums, it is worth noting that it's home to over a dozen archival repositories and a system of 20 branch libraries. So much of my interest in getting involved with Wikipedia can be explained by this chart. The light blue bars represent monthly visitors to wikipedia.org and the dark blue bars represent monthly visitors to all of the Smithsonian's websites combined. As an information professional, I am dedicated to helping people find the resources that they need. Working on Wikipedia helps me to reach as many information seekers as possible. When I joined the Smithsonian in 2007, staff had already been experimenting with editing Wikipedia over the years with mixed results. Since many of us didn't really understand the norms of the community, we often had editors get blocked for conflict of interest. We were even accused of spamming Wikipedia with links to our websites. But in 2010, some volunteers from the DC Wikimedia chapter made contact with some Smithsonian staff members, and together they organized the beginnings of the Smithsonian Institution Wiki Project. It kicked off with an in-person training session for 30 Smithsonian staff where we were introduced to the norms and best practices of the Wikipedia community. This event was the first time that I really grasped the passion and the commitment of the community behind the online encyclopedia. I met a number of Wikipedia editors that day. One of them, Katie Philbert, paid a follow-up visit to the Archives of American Art to speak with me and my supervisor about ways that we could work together. In the spring of 2011, Katie introduced me to another Wikipedia name, Sarah Sturch, who was a grad student in museum studies at that time. And I encouraged Sarah to come an intern at the Archives of American Art in order to help us get more involved with the Glam Wiki Project, and that is how we found ourselves with a Wikipedia in residence. Our goals for our Wikipedia in residence were pretty modest. We wanted some training and mentoring. We wanted to see more of our stuff on Wikipedia, and we wanted to get to know the community. We were not really sure what to expect, but we were open to new ideas, and luckily Sarah had many. I credit our Wikipedia in residence with helping to turn me into a much more savvy Wikipedia editor and advocate, and with helping our staff understand how institutional contributions fit with the Wikipedia platform and community. She also connected us with technical experts who were able to automate both data analysis and image contributions by doing some custom scripting, which is incredibly helpful. She helped us organize backstage pass and edit-a-thon events, really fun and inspiring in-person events where experts and people who are learning got to get together and work on articles. Our events have been attended by very smart people from a range of ages and backgrounds, all volunteers. The volunteers get a behind-the-scenes peek at our collections, and they get to have a little Q&A time with staff. They always come away inspired to edit articles related to our collections. There's a great spirit of learning and curiosity in these events. And even though Sarah Sturch has now moved on to other projects, our organization now has a number of ongoing relationships with a core group of volunteer Wikipedia editors that we continue to work with on a regular basis. So after she finished working with the Archives of American Art, Sarah went on to serve as Wikipedia in residence at the Smithsonian Institution Archives for an additional semester. By the time she left in 2012, an internal Wikipedia community of staff librarians, archivists, and museum folk had emerged. Since that time, there have been two subsequent edit-a-thons at the institution at the American Art Museum last August and a Wikipedia Loves Libraries event at the Smithsonian Libraries this past October. Wikipedia Loves Libraries, which you see here, was our largest and best attended event yet, with around 40 people in attendance. It took place in the main library of the National Museum of Natural History. We even got a small grant from the DC Wikimedia chapter to help provide everybody who attended with a free lunch. The attendees were equally impressed by our collection of rare books and rare akinaderms. One of our greatest successes, in my mind, besides connecting with the community, is our foray into putting our public domain images on the Wikimedia Commons. All Archives of American Art contributed images include a custom template with our approved metadata and multiple links pointing back to our website. After we uploaded our first batch of images, we didn't really do anything else to work on them, but I'm happy to say that the Wikimedia volunteer community set right to work on their own. Hundreds of our images are now being used to illustrate numerous articles on Wikipedia sites in 25 different languages, from biography pages to articles on life drawing and art education to history articles. Let me share one example. Here is the Archives of American Arts record for one of the images that we contributed. It's a WPA photograph of the Harlem Renaissance sculptor Augusta Savage. Over a 90-day period, according to Google Analytics, this page on our website had 29 page views. Could be better. But the Wikipedia article on Augusta Savage was viewed 2,706 times in the same 90-day period. And what does this article feature front and center? The photograph of her that we contributed to Wikimedia Commons with our repository citation, metadata, and multiple collection links close at hand. At this point, the Archives of American Art is getting a sizable amount of web traffic from the Wikimedia family of websites. For the past few years, Wikipedia has been our single biggest referring URL and is responsible for about 5% of our overall inbound site traffic. By comparison, Twitter brings us about 2% of site traffic and Facebook 1%. Given the relatively small amount of maintenance energy that we put into Wikipedia, given especially when compared to the upkeep required on our social media accounts, Wikipedia gives a very nice return on our investment. But even if we didn't get the visitor traffic back to our primary website, it seems to me that diffusing our digital assets to a greater audience on sites like Wikipedia and the Wikimedia Commons is the right thing to do. Isn't this why we digitize our collections after all, so we can share them with the broadest possible audience? Thank you, Sarah. I'm going to pick up the story here at Back at OCLC and tell you about our experience with our Wikipedia and residents. We were really interested in having a residency that focused on two dimensions. One is looking at the special role that library data could play within Wikipedia. OCLC obviously has a stake in interest and expertise in dealing with library data, so we were really interested in exploring those avenues and seeing what special role OCLC could play in this. And also looking as a nonprofit that touches many thousands of librarians worldwide, we wanted to see what we could do to develop a program that would have an educational and outreach component for librarians. So talking to librarians about Wikipedia and telling them, spreading the word more about how they could get involved, talk to them about the GLAM Wiki Project and talk to them about various ways that they could get involved and really heighten their awareness and spread some of the things that we were learning around in the community a little more. So I'm going to talk about our first project and the project that Max has mainly been involved in, which is looking at what we've called VIAFBOT. VIAF is the Virtual International Authority File. It's a project that is managed by OCLC and it is a merger of 20-some national authority files from many countries. It's multilingual, it has broad representation and really is a place for authority files to be merged. One of the things that I discovered, much to my surprise when I talk to people from the Wikipedia community, is that disambiguation is really important and disambiguation of names is really important to them. And to give you an idea of how this is handled in Wikipedia, let's take a look at this disambiguation page for John Adams. So John Adams is many people and it's important to know which John Adams you're looking for. If you search for John Adams on Wikipedia, you may get to this page disambiguation. It starts with politicians, goes on to composers and goes down even quite a ways below this. You can scroll down, down, down on this page. There are many John Adams. So there was really a growing recognition of the importance of library name authorities and Wikipedians were aware that library name authorities were out there and that they were valuable. And some Wikipedia articles had started incorporating what they call authority control into their articles. So not too surprisingly, the German language Wikipedia had already started to add in authority control. And within the German language Wikipedia, I think several thousand articles already had authority data in an authority control template. But the English language Wikipedia had only about 4,000 articles with authority data. So let's take a look at VFBOT. Let's take a look at the authority control template. So here's the lengthy article on John Quincy Adams, the sixth president of the United States. And if you scroll all the way to the bottom of the page, you will see this little template, authority control. And this is where Wikipedians were starting to put links to various name authorities in national files. So we were asked to help out on this. Here is VF and John Quincy Adams records. So if you followed that link, you would wind up here on John Quincy Adams VF page. So VFBOT has now edited over a quarter of a million articles, which now have library authority data in them and links to VF. This is out of roughly a million articles that are about people on the English language Wikipedia. So you can see that this project has already had a pretty significant impact on Wikipedia. And indeed, we can see people now following those links from Wikipedia through to VF. Well, this sounds like a pretty simple thing. Like we just threw a switch like our Wikipedia and Residence Max just threw a switch, and the magic happened. But in fact, there were many, many, there was the approval process for getting a bot or a robot to do all of this editing programatically was quite slow. The Wikipedia community uses bots or approaches the use of bots quite cautiously. They have a long standing tradition of creating and editing articles by hand. Almost everything that you see on Wikipedia is done by humans who care very much about what they're doing and how they do it. So the approval process was necessarily slow and took a great deal of caution. So it went in many stages. First, there was some discussion online in a place called the Village Pump. Then there was a formal request for approval for the bot once it had gone through the discussion stage and people had generally agreed that it was a good thing. Then there was testing and more testing and then running the bot in stop and start mode along the way Max discovered that there were bugs in the bot. And as it ran across Wikipedia, it would run into those bugs and have to be stopped and fixed sort of midway. So I'm going to take you through some of the steps. And of course, the wonderful thing about Wikipedia is that everything happens on Wikipedia and you can view all parts of this. So if you're curious, you can go to the Village Pump and view the discussion on authority control integration. And you can see that back in June, this discussion started. And for the most part, this discussion was quite positive. There were a few people who were truly skeptical about this. People asked questions, very good questions. And the discussion here really helped to shape the bot creation. So it wasn't just something that Max cooked up and then went forward with. It was really shaped from the very beginning by this discussion that happened on Wikipedia. The next thing that happened was, as I mentioned, a formal request for approval. And this too had discussion and it laid out how the bot was going to proceed. And it was approved, as you can see. But again, you can go see all of the discussion there. And then once it had approved, Max was able to create an account with the username VFBot. And you can go to the VFBot user page and look at the VFBot information about VFBot. But furthermore, you can go to the talk page on VFBot and see all of the terrific feedback that we received from the Wikipedia community once the VFBot started to run. And all of the really great and concrete suggestions that we got, alerts when it wasn't working correctly, et cetera, that went on. So if you're curious about that, you can go to the VFBot user page. I'm going to switch now from talking about library data and talk about education and outreach because we think that as a library services organization at CLC, has the ability to reach many, many librarians and that we can do so at scale. And what we've really been doing is focusing on, we've had a series of webinars focusing on what we've called Wikipedia Literacy for Librarians. We had two separate webinars in July and August. One was targeted at research libraries and pushed out through OCLC research. And with the other webinar, we targeted public libraries working with OCLC's web junction, which has fantastic connections within the public library community. And there were over 500 librarians that attended the live events, so the live webcasts. And we know that these webcasts have been viewed after the fact, but just the fact that we were able to put on two webcasts and reach over 500 librarians was really great. We've also done some in-person events. We had a Wikipedia Loves Libraries event at Wikimania, which was held in Washington, D.C., during July. And filled a room, I think we had probably at peak around 100 people attending this. And this was a mix of Wikipedians and librarians, pretty easily evenly split. A lot of energy and excitement about Wikipedia collaborating with libraries. And a lot of support for our idea of working with library data in Wikipedia. We also experimented with a MOOC, a particular MOOC on the platform called P2PU, which is peer-to-peer university. This was themed around Open Access Week, but it's still available. Anybody can go and take it. And this is a six or seven-part class that you can take in components. And when you complete it, you can get a badge on your Wikipedia user page, saying that you've completed this Open Access media challenge. And this is, again, to help introduce librarians to the concepts in Wikipedia. And learn how to do a little editing, learn how to create a user account, and become acquainted with some of the concepts behind Wikipedia. Switching over to the results that we have had in VFBOT, you can see the lines at the bottom of the page are, it's kind of hard to see on your screen, perhaps. But this shows overall inbound traffic into VF, the virtual international authority file. And the purple line that you can see kind of hanging out in the middle there is the German language Wikipedia. And I mentioned a while ago that the German Wikipedia's have for some time been adding in links to library authority data, and VF was one of their sources. But right around the time that we started running the bot in October, you can see the English language Wikipedia coming in and actually lifting the curve of traffic coming into VF in a noticeable way, and then continuing right on through into the end of November. We're not sure if this is a trend that's going to hold if these are Wikipedia's who are curious about what these links are and following them through. But we think that this is probably a trend where people see the links at the bottom of the page or curious about them for one reason or another and may follow them through. I'm going to talk about kind of a more tangible sort of result. And this is a story that we heard from a cataloger named John Myers in Union, who works at Union College and connected to New York. And he wrote to us back in October to report, I had an Arabic name to enter into a record as part of a note and I wasn't confident about the diacritics. So I looked in an authority file to temporary download it, copy the file of the name and move on. And he couldn't find the name in OCLC because again he wasn't sure of the diacritics. So he looked it up in Wikipedia under his common name and bingo! He found the guy in Wikipedia, so here's his article. Even better, he continues, Wikipedia has a link to VF, double bingo. With the authorized form of the name in VF, I could readily find the record in OCLC. The miracles of an interconnected bibliographic dataverse. Thank you for your efforts. It made my work in my catalog possible. But what's really nice to us is that we got this note from him, I think, on October 12th. And when we looked back, VF Bothead added that VF link just days before. So this is a real, a fine example of how VF is working to actually help not only Wikipedians to send big U8 names, but also to help with catalogers. So looking for, oh, and here is that individual's name in VF and his VF record. So we think that there are more opportunities to extend our Wikipedia Loves Libraries events Here is a picture from an event we had just weeks ago at the Seattle Public Library. We think that public libraries are probably great spaces to bring together Wikipedians in person together with librarians. So here we had, we brought together librarians from the Seattle Public Library. A small group of very experienced Seattle Wikipedians who are really interested in working on documenting local history and culture in Seattle. Together with some people from the University of Washington who work in the University of Washington Libraries and Archives who were quite interested in learning more about Wikipedia and more about the role that their resources might play within Wikipedia. It was a small event, as you can imagine, something on a Saturday in December might be. But we had really great energy in the group and people made a lot of good connections and we're going to be following up to see how we can help encourage this group meeting and think about how we can scale out these types of meetings going forward. We also think we could perhaps develop a best practices document for librarians adding content to Wikipedia. Other suggestions we've had are how can you find Wikipedians within your local community if you're starting from scratch? How do you make those connections and contacts? So those are all things that we're thinking that we could probably help with providing some documentation and some guidance for people who want to get started. So there's certainly a lot going on there. There's also additional opportunities we think for library data. So you probably see these little sidebars within Wikipedia. They're called info boxes. If you look on an article that is about a book, you will see a template that's called info box book and it contains fields, optional fields that look a lot like metadata. So we'd like to put in ways, an OCLC number perhaps maybe help to populate these info boxes based on WorldCat data and put in easy ways for people to get to that book in their local library, which is certainly something we can help with. And I think again just making sure that library resources and the ability to get to library resources is highlighted within Wikipedia and for Wikipedians is a very positive thing and something that we can help with. So we'd like to thank you very much for your attention. Sarah and I would be very delighted to get your questions. You can feel free to email us or get in touch with us on Twitter. We hope to hear from you and thanks for attending this presentation.