Trip Report of DPLA Audience & Participation Workstream

On December 6, 2012, the Audience and Participation workstream met at the Roy Rosenzweig Center for History and New Media at George Mason University. About two dozen colleagues participated in person and remotely via Google+ Hangout to talk about processes and strategies for getting content into the DPLA (the content hubs and service hubs strategy), brainstormed on the types of users and the types of uses for the DPLA, and outlined marketing and branding messages that aligned with the goals and technology of the DPLA while getting content contributors and application developers excited about what the DPLA represents. I’m happy to have been invited to take part in the meeting, am grateful to DPLA for funding my travel to attend in person, and came away excited and energized about the DPLA plans — if also with a few commitments to help move the project along.

Emily Gore, DPLA’s Director of Content, started the first topic by describing the mechanisms being set up to feed metadata to the DPLA database. The first version of DPLA will be an aggregation of metadata about objects various services and cultural heritage organizations around the country. The DPLA will leverage and promoting metadata coming through hubs, where a hub can be an existing large gathering of stuff (“content hubs” — think Harvard University, Smithsonian Institution, National Archives and Records Administration) or a hub can be a meeting point for state or regional content (“service hubs”). From the perspective of the Audience and Participation workstream, the service hubs are probably the most interesting because that will be how information about an institution’s content gets into the DPLA.

Just about every state in the country is covered by a state or regional digital library program, so the infrastructure is already out there to help organizations. The DPLA itself is aiming to be a small organization of about five to ten people, and at that scale it would be impossible to have a one-on-one relationship between the DPLA and all the possible organizations in the country. So the DPLA Service Hubs will offer a number of services in a region: aggregation of metadata from local repositories, help with new digitization and creation of associated metadata, and engaging participants in the region around different uses of the content in the DPLA. By the April 2013 launch of the DPLA site, the goal is to have seven service hubs operating and a similar number of content hubs. Some of the state and regional collaboratives have already reached out to the DPLA to join, and DPLA is working on updating a list of collaboratives that was created a few years ago. One path of outreach is through the Chief Officers of State Library Agencies (COSLA) group. Talking to state library agencies makes sense because there are indications that IMLS — who grants money to state library agencies — is aligning its LSTA funding with the goals of participating in DPLA. State humanities councils and ALA can also be venues for getting the word out. The ALA Washington Office can be especially useful for getting word to legislators about the importants and value of collaboration with the DPLA.

We talked about how there are technical tasks involved with adding new hubs to the DPLA — it isn’t as easy as just ingesting and indexing new metadata. There will be iterations of mapping adjustments, tuning the weighting of fields in the index, and other tasks, so DPLA will need to set expectations about how fast it can add new hubs to the service. It was noted in the meeting that the service and content hubs will in one sense be customers of the DPLA and in another sense will be providers to the DPLA. This relationship will be akin to a business-to-business relationship, and it will be important that the DPLA can provide adequate “customer support” to the hubs to make the relationship work out best.

The focus at launch is on cultural heritage objects, books, and manuscripts. The common denominator is that the metadata must be sharable under a Creative Commons Zero (CC0) license, allowing for the free reuse and remixing of the metadata. In this form, the DPLA will be an index of descriptive metadata that leads the searcher to where the item is stored at the contributing institution. That institution can specify other rights on the reuse of the digital object itself. Interestingly, the CC0 policy for metadata is a source of concern for some potential DPLA participants. Where libraries have less of a sense of ownership over the metadata describing their objects, the museum community has a higher sense of ownership because of the extra effort they put into creating and curating this metadata.

We talked for a bit about the impact that the visibility of DPLA will have on desires for organizations and even individuals to digitize, describe and electronically mount their content. (“If they have stuff like that, I have stuff like that, too, that I want to add.”) The DPLA can be helpful by providing clear pipelines for the processes for content to be added to places that will be harvested and integrated into the DPLA. Perhaps even bringing digitization “to the masses” by going through the local historical societies where there will be opportunities for conversation about what is good to keep and how to do it.

This discussion of what content will be in the DPLA lead into talks about the kinds of people using the DPLA and what they will want to use it for. The goal is to create “personas” of DPLA users — fictional representations that encompass research about the users, their motivations, and their desires. (As examples, we briefly looked at the HathiTrust personas and the earlier work on DPLA Use Cases.) The driving goal is to give these personas to the contracted developer (iFactory) for use in creating the initial front end website. As an aside at this point, the heart of the DPLA at this stage will be the aggregation, enhancement, and serving of descriptive metadata to software applications that remix and display results to users. One way, but not the only way, this will happen is via the http://dp.la/ website interface being created by iFactory.


We brainstormed the possible labels for personas: Casual Searchers, Genealogy, Hardcore Enthusiasts, Wikipedia / Open Source Folks; info nerds, Small business / startups, Writers / Journalism, Artists, Students, Public School Teachers, Home schoolers, Scholars, Other Digital Libraries, State Libraries, Public Libraries / Public Librarians, Museums, and Historical Societies. We also brainstormed a whole slew of behaviors that these personas could do (several hundred post-it notes worth), and then grouped them into broad categories:

  • Finding Specific Knowledge: school research, curricular-related; local/personal history; specific “laser-like” focus; open-ended, on-going activity; awareness of a body-of-knowledge problem
  • Learn: skill-acquisition (things that take longer, as a project)
  • Harvest and/or reuse: visualizations, build new collections
  • Contribution: contribute content; enhance metadata (DPLA needs to be able to answer the question “I want you to add X”)
  • Sharing/Connecting: outwardly-focused; using DPLA as a point to go out to other people, find partners, start a book group, sharing something cool with, friends; building community; connecting institutions, see what other libraries are doing, sharing content with other libraries
  • General, accessibility: featurish-type notes

After a little more refinement in sorting and labeling, these behaviors will then be used to create the characteristics of the personas.

The last activity was talking about branding and marketing — how to get organizations and individuals excited about using the DPLA. A backdrop of this discussion is making people — especially funders — aware of how DPLA is an enhancement to every library’s services and collections, not a replacement for them. That the DPLA will be seen as complimentary to the local library came out strongly in the October DPLA Plenary session. Among the discussion of “what’s in a name?” (‘dp.la’ or ‘library.us’ or something else) and what is it that DPLA wants to pitch to users (the metadata platform, a single-entry homepage at http://dp.la/, or an app store of DPLA-enabled application), was a fascinating discussion about getting developers interested in the DPLA platform and programming interface. We talked about getting library school, college, and high school classes interested in building DPLA apps as term projects. We also talked about the role of existing organizations like Code4Lib and LITA in introducing and supporting developers creating applications using the DPLA platform.

In the end what emerged is a possible thread of activities from the midwinter meeting of the American Library Association (ALA) through the Code4Lib conference into South-by-Southwest and the annual ALA meeting. The thread goes something like this:

  1. Petition LITA to form an interest group for DPLA activities at the ALA Midwinter meeting, and possibly hold a forum there for interested librarians.
  2. Hold a half-day preconference tutorial at the mid-February Code4Lib meeting in Chicago covering example uses of the DPLA API, effective ways to process and remix JSON-LD data (the computerized format of information returned by the DPLA platform), and discussions of the content available in DPLA.
  3. Use the Code4Lib meeting to launch a four to five-week contest for teams of developers to create interesting applications around the DPLA platform.
  4. Show the entries at an already-arranged presentation at the South-by-Southwest conference in mid-March, and announce a winner.
  5. Arrange for space at the ALA Annual meeting in June in Chicago for a more in-depth discussion and exploration of DPLA.

The hook for developers is showing them a new, vast store of liberated data that they can use to remix with other data, create absorbing visualizations of the data, and facilitate user engagement with the data. The DPLA is going to become a huge set of liberated data, and we think that can be attractive to not only library developers but also developers outside the traditional library and cultural heritage community.

And with that we ended the meeting at George Mason University. As I said in my previous post recounting the November Appfest meeting in Chattanooga, these are exciting times when the reality of the DPLA vision can start to be seen. I’m eager to see, and participate as much as I can, in the effort.

The text was modified to update a link from http://dp.la/get-involved/events/dplamidwest/ to http://dp.la/info/get-involved/events/dplamidwest/ on September 26th, 2013.

The text was modified to update a link from http://dp.la/workstreams/audience/ to http://dp.la/wiki/Audience_and_Participation on September 26th, 2013.

The text was modified to update a link from http://dp.la/about/digital-hubs-pilot-project/ to http://blogs.law.harvard.edu/dplaalpha/about/digital-hubs-pilot-project/ on September 26th, 2013.

(This post was updated on 26-Sep-2013.)