MM: Curation spins off of collection development. It’s about arranging a set of resources that meet clearly articulated standards, around a certain theme. It’s the ways in which collection is promoted. If we focus on creation and sharing, the promotion piece is the extending piece, the piece that allows people to take it further. Content wasn’t as accessible in a less digital world.

MM: I was a practitioner originally. I worked at a science-focused private school in San Antonio, so science was my entry point. Later, I worked on the Last Mile Project, an NSFNET partnership devoted to building national backbone network service. Then I got an NSF grant and some Michigan Department of Education money to found a digital library for teachers. We set out at that time to build a highly curated collection of resources. There were fewer resources to choose from then, and we came upon some early issues relating to curatorial policy that curators had to deal with. For instance, whose notion of quality are we really putting forth?

We went through a couple of different models of who would be involved in our collection development. We weren’t 100% satisfied with the stuff selected by the folks who were just classroom teachers because that represented a certain perspective. We weren’t 100% satisfied with the way the collection development looked with people who were solely domain area experts because that represented another perspective on what the resource is about, what it’s used for, and how to promote it. Ultimately, we decided that our collection developers, or curators, should be school librarians who collected in their area of domain expertise. So, for instance, if you were a K–5 school librarian with a biology degree, you would be collecting in K–5 science. That represented the best of all worlds.

We got the pedagogical piece, we got the domain expertise piece, and we got the other domain expertise piece. We needed to consider: What does a high quality information resource look like? For instance, is it persistent? Is it usable by multiple populations?

It was my first entrée into taking what I knew to be best practice and what we saw evolving as best practice. It was my first entrée into what does it mean to take what you know as collection development and to put it outside you as a policy? We developed a set of policies because we wanted to ensure metadata quality as well as resource quality. I did that from 1998 till 2008. The Michigan Online Resources for Educators [M.O.R.E.] still exists.

We were one of the first digital libraries to assign curriculum standards to digital resources, so we wanted to make sure we were creating the highest quality metadata possible and not let it become a complete free-for-all. But certainly users could suggest, tag, rate resources.

Marcia believes that user-generated content and commentary are valuable, but first-level cutting is important in education resources.

MM: The user voice has to be there, but there are so many different views of quality assessment. As the person running a collection, you need to stay committed to the idea that you have to decide what standards you are going to set and how are you going to enforce those standards. That’s the whole difference between curation and the free web. What you’re promoting is your view of how things need to go to together around a particular theme and the best resources to reinforce that theme. You need to make choices; you need to clearly articulate policy. The people who were our collection developers out there in the field believed in our policies.

MM: It’s important to curate. I have a hard time seeing a place where an information professional wouldn’t have a lot to contribute to the curatorial enterprise.

My work centers around open education resources. Learning takes place best from high quality contextualization of resources. While I am not against the British Museum model, that has to be highly mediated; it is all about the context you wrap around it.

At the National Science Digital Library [NSDL], there was a strong commitment to letting a thousand flowers bloom—let everything in, let the user contextualize it. But what we were hearing from users was: “If I am going to go to a curated collection, or a collection that seems like it should be curated, I expect curated stuff to be in there. So, if you’re going to collect stuff, and you are not going to assure to me that it has good science, that it’s appropriate for the audiences that it’s intended for, we’re not interested.”

That might be the role of metadata. That might be the role of the person presenting the collection. In each one of the roles, I see a place for the voice of an information professional.

In a formal education situation, with agreed-upon standards of knowledge like good science, there is a lot of discovery to be done, but there are some foundational elements we need to at least put in place.

MM: This a very interesting time in the lives of all of these projects. They are transitioning into a post–federal funding period. It makes sense to pool the curatorial piece and the policy and philosophy pieces. It’s not about building, it should be scalable and demonstrate efficiency in management.

Everyone is pleased to swap technology approaches. I think there’ll be a larger push for metadata interoperability, to share objects across collections, and exciting approaches to search. We may see a metasearch develop. From the user perspective, perhaps the pesters from collection to collection will disappear. Perhaps search will become more seamless.

There are territory issues. Projects want to own their individual identities. Trying to get people to agree upon a common metadata standard is never really going to happen, and maybe it shouldn’t happen. However, better crosswalks and better interoperability strategies can address a lot of issues. However, anytime you create a crosswalk, you run the risk of losing something. This is a role for the information professional that I see persisting.

MM: There are new opportunities for tool-builders to curate so all the user sees are the objects that are dynamically combined from across collections. We’re moving toward personalization, and a spot for such a tool will not be left empty for too long. So it will become less about visiting the individual collection and more about the tool learning what you need or you telling the tool what you need and it pushing the resources to you from trusted providers.

Marcia Mardis and her team have developed a Web2MARC tool to facilitate the import of web-based OER content into local catalogs (figure 5.1). I asked her about the progress of that project.

MM: Right now we’re building out the functionality in a number of different areas. For K–12 users, we’re building out the next generation science standards applications, the Common Core State Standards [CCSS], some additional vocabularies, controlled vocabularies—there are already uncontrolled vocabularies features. We’re building additional capacities for ratings, reviews, and other aspects of personalization. The other thing we’re looking at is increasing the number of output formats, not just MARC, but XML and a few other formats that will make it easier for the records to be ingested in a number of different types of repositories, not just the school library catalog, but also, for instance, the state Department of Education’s OER repository. You can already bring records into your personal repository like Zotero or Endnote.

We’re also spinning off an API part, a plug-in so you can pin records on Pinterest, share on Facebook, and tweet. You’ll be able to click on a button and generate a record from wherever you’re browsing, rather than having to copy the URL, go to the Web2MARC tool, paste the URL in the box, and create the record.

MM: Yes, but it will result in a record you can take with you and put wherever you would like it to go. The nice thing is that every record generated goes into a common database so that as a user, you can grab a record out and it becomes its own self-curated depository of high quality descriptions. As people refine those records, it becomes a crowd-sourced repository of high quality descriptions. And then repository providers can harvest that repository as well, and it becomes part of a life cycle of high quality metadata.

MM: I see opportunity everywhere. I am just really excited. There are so many people in libraries who have been shaming themselves for really enjoying building their collections. That’s just too bad. It’s such a fantastic and singular activity. It’s at the core of what we do, and the rest of our jobs do not make sense without it.

Digital Libraries to School Libraries Project [DL2SL] Web2MARC Cataloging Tool

http://dl2sl.org/about/project

Michigan Online Resources for Educators (M.O.R.E.)

http://more.mel.org

National Science Digital Library (NSDL)

https://nsdl.org

DJ & MJL: Not an easy one! We believe that digital curation is synonymous with digital preservation. We have a collection development policy that guides identification, selection, and preservation of our unique collections.

We tend to use digital preservation. We believe in offering multiple access points to our collections that comprise scanned and born digital documents and resources. The collection is available from the Maryland State Law Library Catalog and from WorldCat. We harvest documents, taking the entire content and preserving it in PDF format. We have found that, over time, web information disappears. Our annual survey data show that 50% of the digital content from .gov sites has disappeared.

DJ & MJL: We use a variety of tools to find content of interest to judges, lawyers, law clerks, and law students, such as Google Alert. We outsource the scanning of print documents. PTFS Archivalware allows us to store, search, retrieve, browse, and manage a diverse set of content/digital objects within one easy-to-use system. Content is cataloged and organized with CONTENTdm Digital Collection Management by OCLC; editing metadata; and index documents.

DJ & MJL: No one else is going to make these valuable resources accessible. We are earnest at preserving information; we use the analogy of “saving baby seals” as endangered species.

DJ & MJL: Debbie, as Maryland collections librarian, is responsible for harvesting information resources. Mary Jo is head of collection management.

DJ & MJL: We get great feedback from the Rules Committee Materials reports—we care a lot about this section. In addition, we get anecdotal information from our walk-in clients.

DJ & MJL: We hope to expand our collections as time and financial resources are granted. There will be a new collection accessible soon, for instance, the 19th Century Maryland State Bar Association Transactions Collection (late 1800s to 1990s). We are looking at expanding marketing our collections. Our director Steve Anderson has a forward focus for collection development.

Maryland State Law Library

www.lawlib.state.md.us

Collection Development Policy for Maryland State Law Library

www.lawlib.state.md.us/aboutus/policies/CollectionDevelopmentMSLL.pdf

Maryland State Law Library Digital Collections

www.lawlib.state.md.us/collections/digitalcollections.html

“Link Rot” and Legal Resources on the Web

http://cdm16064.contentdm.oclc.org/cdm/reportsandpublications

LF: It’s hard to ignore the museum definition of curation, having worked in museums and having taken a course in curation. The core piece of curation includes the responsibility for a collection—what to collect, how you document the acquisition process, registration, cataloging, displaying, loaning and maintaining, consideration for provenance, doing interpretation, supporting documentation, looking at context and significance, sharing that information with the public, as well as display in whatever format. It’s a cluster of different functions.

LF: Even in a completely digital museum, you are still talking about selection of recorded information in its broadest sense. There are some philosophical issues for consideration. And there are two ways to dance that out. You have straight representations of knowledge—blog, digital art, recorded song. And then there are reproductions of artifacts, like a photograph of an ancient pot. One is the actual item; the other is a manifestation of the item. There’s a difference.

With a photograph of a 3-D object, the camera angle, the lighting, all that stuff, impacts it. And then we need to consider, what is your definition of digital and where do 3-D printers fit? It’s very complex. Very nuanced, very exciting.

If you are looking at digital artifacts, you are still talking about how they were created, who created them, where they are housed, what is the affiliation. You’re talking about Dublin Core stuff—about description and context.

When you put together displays and portals of information, you need to consider what kind of container will you choose. How will you present? It’s not core librarianship, but it’s an aspect of our work that’s becoming increasingly important.

When we’re teaching whatever level, it is important to think about information as a product; it has a container of some sort. We should be thinking about the context, keeping in mind the objective, the audience, the message we want to convey, what container we will use, and how will we share it. It’s very intentional. You bring your own interpretation. It’s also a relational thing.

LF: Yes, if you’re doing active curation, you need to be aware of voice. When we’re teaching kids to curate, we need to ask them about their voice.

LF: In that respect, museums are intentional in their development. Institutions have their missions. Collections represent the institution’s perspective. Kids need to be aware of that. Whether it’s a museum, your public library, Pinterest, or Digg. Selection is a very human, very intentional endeavor.

LF: Those exhibits only represent 5–10% of the collection. Right. There’s the big collection, and then there’s the educational aspect. The curator determines what’s educationally important, of value. And then there’s the interpretation that involves the story, the displays, the sequencing.

DC: The DPLA is truly a social project and a political success, achieving buy-in from growing numbers of stakeholders at a variety of institutions. DPLA is a complementary service to these digitized collections by providing discovery layers for America’s collections. I’m thrilled about the current status. We have materials with over 400 languages represented in the currently accessible materials, and several more digital libraries are interested in partnering with the project.

DC: There are many different models out there, but the goal was not to rely on specific collections, but rather on how many different kinds of collections can come together and bring them to one place. So the model is a hub networked one. The central office doesn’t host, but instead links out to content hubs. It’s a very webby model.

DC: DPLA does not currently have a formal agreement in place with the Library of Congress, but we are hoping to eventually come to an agreement. Right now, the main focus is “surfacing content” from all over the country.

DC: The main thing is completing the map—getting content hubs in all 50 states. These are “on-ramps” for collections residing in large and small archives and libraries to come on board. We’re planning for addressing subject and chronological and format gaps. Also, usage. We’re thinking about ways to bring DPLA into classrooms and community programs.

DPLA’s community representatives program is likely to be a major tool for meeting some of these needs. Currently, there are over 200 volunteer reps from all states and some foreign countries spreading the word about DPLA at the local level. DPLA does not hold the collections, but rather the metadata that enables seamless connection to resources from this one-stop portal. Building off of existing infrastructures, the goal is to be complementary to existing collections and to work with small cultural institutions to help their materials see the light of day.⁹

Luis Herrera, DPLA board member and city librarian of San Francisco, believes that removing barriers to access to this broad and growing range of culturally diverse materials across the country will be a key to the continuing success for the DPLA.

LH: It’s about sharing our cultural narrative. DPLA is expanding access for the nation and beyond for rich content and culture at the local level. We’re broadening reach and scale. We’re excited about bringing these amazing collections to a much more accessible, user-friendly portal.

LH: The initial vision was to have content reflect the broad range of the country. With over 7 million items and over a million visits to the portal, we’re making great progress with content and service hubs. It’s come a long way. Six months ago, California was not that active, so we saw this need to push from the grassroots to have the state library become a content hub and establish a new pathway.

LH: The California Digital Library model brings public library engagement in academic collections, a hybrid model. The success will be in how we scale it and how public libraries can also provide content services and build capacity. The local libraries prioritize collections and determine how to share with content hubs.

In San Francisco city libraries, space has been repurposed as digitization centers, inviting community members in to document and scan collections of small neighborhood archives. The digitization program not only brings attention to these small collections, but also provides training to those who want to learn how to digitize materials, exemplifying the makerspace model. Initiatives such as San Francisco’s align perfectly with the personal digital archiving outreach program of the Library of Congress and reinforce the importance of archiving neighborhood and personal histories. In this way, DPLA can fill gaps for communities that have not yet been represented.

LH: It is a democratic, inclusive approach, with people at the grassroots level determining what is of value and what needs to be shared.

LH: We need to find a way to engage younger generations and find ways to engage a younger demographic, create some involvement in schools. We want opportunities for students to get involved and share their own unique content. It represents changes in how we are learning.

What is striking about the DPLA is its capability to be leveraged not only for research and scalable sharing of resources but also for education in curation for young people. As schools increasingly extend learner ability to go beyond information consumption to content creation, DPLA could potentially see escalating involvement from communities of young learners, matching DPLA’s goal of moving beyond geography to include learning communities of all types.