DPLA and the International Image Interoperability Framework

By Mark Matienzo, May 10, 2016.
Published under:

350x323_IIIF-logoDPLA, along with representatives of a number of institutions including Stanford University, the Yale Center for British Art, the Bibliothèque nationale de France, and more, is presenting at Access to the World’s Images, a series of events related to the International Image Interoperability Framework (IIIF) in New York City, hosted by the Museum of Modern Art and the New York Academy of Medicine. The events will showcase how institutions are leveraging IIIF to reduce total cost and time to deploy image delivery solutions, while simultaneously improving end user experience with a new host of rich and dynamic features, and promote collaboration within the IIIF community through facilitated conversations and working group meetings.

The IIIF community provides the following overview for its mission and goals:

Access to image-based resources is fundamental to research, scholarship and the transmission of cultural knowledge. Digital images are a container for much of the information content in the Web-based delivery of images, books, newspapers, manuscripts, maps, scrolls, single sheet collections, and archival materials. Yet much of the Internet’s image-based resources are locked up in silos, with access restricted to bespoke, locally built applications. … IIIF has the following goals: to give scholars an unprecedented level of uniform and rich access to image-based resources hosted around the world; to define a set of common application programming interfaces that support interoperability between image repositories; and to develop, cultivate and document shared technologies, such as image servers and web clients, that provide a world-class user experience in viewing, comparing, manipulating and annotating images.

Just like our Hubs are much more than data providers to DPLA, we are also much more than an aggregation. Our core services, which are driven by metadata aggregation, are proving to be successful, indicated by our traffic, with 57% of that traffic via portal and 43% through our API. However, part of our larger role is to stand behind the development of efforts that make cultural heritage materials easier to use and share by anyone who wants to use them. Accordingly, we believe it aligns with our larger mission to support the development and implementation of standards and software that make encourage interoperability and reuse of the materials that we aggregate. In turn, from our perspective, DPLA’s support of the International Image Interoperability Framework aligns naturally with these efforts, and we have been encouraging the adoption across our network.

DPLA has a number of motivations for promoting the adoption of IIIF within our network of partners. As noted, we see a high level of value in the use of open standards both within our community, as well as within allied communities within which we participate. However, IIIF also allows us to begin to address some larger needs at DPLA as well, particularly in terms of improving the user experience of accessing, delivering, reusing, and annotating image resources from our Hubs and partners. Our experience has shown us that this work will also have high value internally at DPLA, allowing us to more easily reuse image content in exhibitions and other curatorial contexts. In particular, we are aware of user experience issues through user testing of the DPLA portal, much of which relate to the “last mile” aspects of delivery to resources that we have aggregated. While some of these issues are not necessarily related to images, these aspects nonetheless impact images for us consistently. Across the board, we have discovered that access to images can often be unclear for many users, especially once they land on a DPLA item page. Furthermore, this is not only true for portal users, but API users as well. A lack of a reliable API to identify images that may provide zoomable views or at specific sizes is essentially impossible right now, without crawling the remote site. The consistency that IIIF would bring to the DPLA community would allow for greater possibilities of reuse.

An image from photographer and moving image pioneer Eadward Muybridge's Animal Locomotion series, contributed by Boston Public Library via Digital Commonwealth, http://dp.la/item/7fba90b480b0bcd8ff414238b4c86773.

An image from photographer and moving image pioneer Eadweard Muybridge’s Animal Locomotion series, contributed by Boston Public Library via Digital Commonwealth, http://dp.la/item/7fba90b480b0bcd8ff414238b4c86773.

Currently, there are five DPLA hubs with production IIIF implementations. Three Hubs, Digital Commonwealth, Harvard University Library, and the Internet Archive, all have production services running. Two additional hubs have implementations of the Image API. California Digital Library’s new version of Calisphere supports IIIF for a subset of images from a number of specific institutions in the University of California system, including UC Riverside and UC Merced. Finally, two additional hubs, The David Rumsey Map Collection and ARTstor, have been working to push their implementations to production.

Nonetheless, there are some issues that serve at least minor barriers to an exhaustive rollout of IIIF at DPLA, regardless of the value or possibilities implementation would provide. First, DPLA needs to establish how to best represent IIIF-accessible resources within DPLA Metadata Application Profile is based. We have been communicating with both the IIIF community and staff at Europeana and the National Library of Wales about potential modeling decisions, and significant progress was made at the IIIF meetings in Ghent, Belgium in December 2015. Secondly, DPLA doesn’t always know that IIIF resources exist for a given item we’ve harvested, often because the institution hasn’t specified this in the metadata about the item. We are interested in hearing from Hubs and institutions willing to work with us to determine a reliable and consistent way to do this. In addition, we are also concerned about the potential user experience mismatch between IIIF-accessible resources and those which are not, and how to best provide guidance on understanding usage statistics for IIIF image access. We hope to address this in conversation with the IIIF and DPLA communities in the coming year. Finally, we realize that IIIF might be a high bar to cross for some institutions, so we have been considering a number of options, including speaking with vendors and possibly providing an IIIF service, to make it easier to expose image resources effectively.

We are enthusiastic about the possibilities, and hope to be able to prototype IIIF implementations with content from DPLA partners in the coming months. We are interested in hearing your thoughts on this, particularly if you’re part of the DPLA network and have implemented or considering implementing IIIF, so please contact us!