 Okay, so we're in the sprint mode here. The second of the beta sprints joins Clear, DLF, IMLS, and DCC, amazing group represented by Rachel Frick and maybe some colleagues coming up on stage. Welcome if you would. There's somebody who has not been mentioned yet but deserves it and then some colleague, Don Waters from the Mellon Foundation who I saw somewhere up there. Don was a founding member of the steering committee and he actually made a grant to Clear and company to develop this beta sprint which was a very helpful thing and I think it speaks to the model we hope to develop for the DPLA that of course the system itself is supported and we need to sustain it much as Europe HANA is but also that grants might be made to partner organizations in order to do things like just what Rachel and crew have done. So over to you. Thanks, thank you for this great opportunity and I'm of course embarrassing all of my fellow colleagues here by making him stand. Sit here while I ramble on for about seven minutes but as John Palfrey says, I am Rachel Frick and I'm the director of the Digital Library Federation Program at the Council on Library and Information Resources and I'm happy to talk about the Clear, DLF, IMLS, DCC, beta sprint full of acronyms, let's go, all right. Basically what we started with is building a metadata aggregation that represented thousands of cultural heritage collections across the United States. We did this by combining some existing data aggregations and data collections, one of them being the IMLS DCC. This was an initiative that began in 2002 and IMLS is you remember the Institute Museum and Library Services. This initiative was started to provide a single point of access to the digital content that was funded by the Institute Museum and Library Services. The other resource that we used was another collection, metadata collection called Opening History. This is an aggregation that's guided by a really clearly defined collection development policy and it builds off the DCC content but it's a base of history collections and extends beyond the IMLS funding content to include other collections and aggregations. We use those two collections of aggregations plus others including things like the DLF aquifer project to build our DPLA prototype. What resulted was a metadata that represented 1,400 collections with over a million items from over 1,200 institutions in 44 states and we were really grateful to COSLA which is the Chief Officers of State Library Agencies to help coordinate some of this content recruitment. As a result, our scope broadened as a result of the variety of collecting institutions that contributed to this aggregation as you can see here. Our work has also been informed by a strong body of research on information seeking in libraries, archives, and digital collections as well as research and metadata quality and collection evaluation. One of the problems with the large aggregation is that retaining that rich collection content and helping users comprehend the sea of items that lie behind a single search box. These are both technical and conceptual problems that grow as your aggregation grows. But as the same time when your aggregation grows you're able to surface new national strengths within this collective collection and bring to surface small collections that otherwise might not be hidden. Within our particular aggregation we had these particular topical strengths including local history of cities and towns throughout the US, military history, in particular Civil War and World War II, arts, Native American history, architecture, women's history, agriculture, and ranching. Our beta sprint prototype allows users to search within a vast array of cultural heritage that has been digitized over the last two decades. Collections and their metadata are key to providing people with intellectual and navigational context, especially as they explore and evaluate materials. Collections in and of themselves are meaningful organizational units that can bring coherence and functionality to a mass of individual digital items. Unlike search engines and retrieval systems that tend to lose this type of collection contents, context, we actually try to exploit that. We preserve that item within its collection context and its geographic context that's often lost within flatter searches. In our beta sprint we also provided an opportunity for users to link to and make use of secondary sources such as Google Books, Hottie Trust, and others. So here is when everybody at the University of Illinois is going to take a collective deep breath because I am gonna do a live demo, here it goes. I really wanted to show you live just some of the great work that was done by our team here. We're highlighting collections up here on this banner screen. Here's a link to our cute little video we made. Over here you can actually see the number of items, collections, and institutions represented in this aggregation. And these are live links so that you can click and get a link to of all these things. You can also navigate by state so you can see we have, Utah we have about 70 collections there, Arizona 84. I don't know what's going on with the Dakotas but we're working on it. I'm doing a shout out to the Dakotas. And also you can browse by date. The colors represent also the density of objects that we have so the more items we have the darker the colors. You can also search by item type. And we've also brought in some other things like entries from our blog which is called Sewing Culture which we highlight different items in the collection. And you can find us on Flickr. That being done I'm gonna go back to, I'm not gonna do search because I only have two minutes. But we really do encourage you to go to our beta sprint site and search for yourself. That being said it's beyond just our portal website, our prototype and the aggregation that was part of our beta sprint. There's also a lot of research, novel and innovative ways of visualizing, analyzing and connecting the aggregations to users. And this is an ongoing process happening at UIUC. One of the examples is the idea of topic modeling to help us refine and evaluate metadata. The other area of interest that I think is really interesting is how do we connect us to the linked open data world? Currently our aggregation already serves as a node to hundreds of cultural heritage collections via collection level descriptions and item level records when available through OAI, PMH. Until larger organizations can provide linked open data the IMLS DCC can help bootstrap larger cultural heritage community by publishing our collection level descriptions in linked open data compliant formats. By connecting these descriptions to existing linked open data sources this aggregation serves as a bridge to collections which are currently hidden from the linked data network. I think this is a great example of a service that the data can do besides being that centralized portal collection. All right, with that we're gonna stop and open it for questions. Here's some URLs. The first one is to the beta sprint. And the one below is to our project website that has a lot of documentation and further information about our activities. And like we said, we wanted to thank Mellon but of course thank the Institute of Museum and Library Services who support for digital collections and the research and the aggregation. None of this work would have been possible. So thank you. Come on, we have all these people sitting on stage. The question is, there are a lot of cultural institutions in the United States. And this is an example, there's a small public library somewhere that's organized by the mayonnaise system. And I'm wondering if you will have a screening process for who is included in your project. So if there's an international association or the United States Association of Crackpots, for example, will you link to them? I'll answer that and y'all can correct me, okay. The IMLS DCC aggregation, it was IMLS funded projects that guide the inclusion there. Opening history has a collection development policy that she can link to. And I think that was very important when you start pulling a huge aggregation, it's what you include and what you're mission and what you're talking about. So there is some guidelines outlined in that collection development policy. Right now the DPLA does not have a collection development policy but we do have a content and scope work stream to help guide what the content is going to be and I think we're gonna have updates from the work streams this afternoon so I'll probably answer your question. I had it back up there. It's actually DPLA for the beta sprint to actually look at the site. It's DPLA.granger, G-R-A-I-N-G-E-R. There it's up there but you can barely read it. .illinois.edu. DPLA, there's a blog site, a blog page that in fact has a list of all the beta sprints that have been made public and we're encouraging everybody to make their beta sprints public. You can find them all at that one spot. Sorry guys, you don't get to answer any questions. There you go. Well you can always reach us by the internet and like you said, on the beta sprint. Thanks.