Photo: Toshio Sato/Tottori University
The United Nations University’s online publications repository, UNU Collections, was officially launched at the beginning of 2014.
Built on the open source platform Fedora Commons with Fez, developed by the University of Queensland Library, as its user interface, UNU Collections provides access to the University’s academic and intellectual output through a single location. UNU Collections is the result of a University-wide initiative that brought together a wide range of expertise from across the University’s institutes and offices to collaborate on enhancing online access to UNU research.
This article shares the story of their progress in developing and delivering this new service to the UNU community, and reflects on the challenges and successes that were encountered over the course of the past two years. You are invited to explore this new and growing database of scholarly works, which will continue to be updated with both archival and newly published material.
The rapid expansion of internet-based information and communications technology calls for renewed efforts by research and educational organizations to better present and disseminate the results of their academic and teaching activities.
As frequent visitors to the UNU website may recall, a major redevelopment of the University’s web presence was undertaken in 2010, resulting in the launch of a newly designed unu.edu in April 2011. Built on WordPress, the new site offered a flexible, cutting-edge and customizable platform that captured UNU’s ongoing transition towards becoming both a research and degree-granting institution.
These efforts were a small-yet-important step in a larger set of initiatives that sought to articulate more effectively and clearly the University’s core activities in an increasingly web-dominated marketplace. As a flagship initiative of UNU’s digital campaign, the success of the new unu.edu paved the way for refocusing attention on bringing one of the University’s most valuable intellectual assets — the corpus of its academic publications — to the centre stage.
Publications at the UN University, as with other think tanks and research and academic institutions around the world, represent some of the primary vehicles for the delivery of innovative and groundbreaking findings to key stakeholders. Through the long and often arduous process of publication, these outputs can be exported to the global marketplace of ideas, where they are further reviewed, interrogated, distributed and debated in the hope of providing an empirical foundation to the formulation of public policy or a theoretical roadmap for the construction of future research projects — all with the goal of advancing human progress through effecting positive, evidence-based change.
In this context, UNU has consistently placed a strong emphasis on strategies and best practices that will increase the dynamic interaction in the world community of learning and research. In order for the University to continue to fulfil the principles of its Charter, its scholars and information and communications professionals must constantly explore new tools and methodologies that will make UNU research accessible to its intended audiences.
The overarching goal of the project that would ultimately lead to the development of UNU Collections was, in simple terms, to increase the accessibility of UNU’s research for its end users, including its students and academics, external researchers and collaborators, the UN system, government policymakers and the general public. More specifically, it was expected that the deployment of a new system and methodology for storing and classifying UNU research would ideally improve the ability of users to search and browse the University’s publications, enhance dissemination and outreach capacity, introduce sophisticated indexing and statistical tools and, in the end, increase the visibility of the entire body of UNU’s work.
On 6 March 2012, a group of UNU personnel from the Office of the Rector, the Office of Communications and the Library met at the headquarters in Tokyo to discuss the current situation of UNU’s online publication presence and to form what would become the core team of the University’s publications working group. The participants in this meeting highlighted what had become a growing concern across the UNU community: from an end-user’s perspective, accessing UNU publications on the web could be a difficult and somewhat confusing undertaking.
Content was distributed over a number of websites, each managed separately and organized in different ways. If one wished to find what UNU had published on human security in 2011, for example, there was at the time no simple way to find this information, let alone obtain access to full reports or briefs from a single location. Taking this perspective as a starting point, the idea of a unified platform that could house and make available all — or at least most — of the University’s publications seemed an ideal solution.
The first task of the team, then, was to investigate potential web systems that could meet this need, define the basic requirements that a new platform should meet, and ultimately test and decide upon which solution to take to the prototyping phase.
Project meetings in April and May 2012 took up the question of how the University’s publications were actually being stored online with respect to the types of databases and classification schemes, and the scope and nature of bibliographical metadata. As was anticipated, this information was in some cases not readily available.
Nonetheless, a broad survey of the University’s online publication presence was undertaken, resulting in the compilation of a comprehensive (albeit relatively superficial) set of data. This initial stock-taking exercise and subsequent analysis quickly revealed the sheer magnitude of the project: in total, there were thousands of publication records spanning 15 different institutes, in some cases over a thirty plus-year period.
From the outset of the project, the support and collaboration of personnel throughout the University was highlighted as a crucial element in the project’s success. With limited human resources at UNU Centre in Tokyo, subsequent phases of the project gradually adopted a top-down, bottom-up development model that had seen success in previous University-wide initiatives.
In May 2012, the Tokyo team submitted to the directors of UNU’s institutes and programmes at the University’s biannual Conference of Directors (CONDIR) a paper on the project that outlined its objectives, its future direction and a proposal for building a coalition of collaborators across the University. Coordinated from the Office of the Rector, a network of focal points at each of the University’s institutes was established with a mandate to liaise and work with the project team to identify institute-specific requirements, share information on the classification of publications, participate in the testing of a prototype repository, and assist in the design of a sustainable approach to maintaining a new, central system.
The project team was consistently determined to deliver a product to the entire University system that was of the highest quality and that would meet the various requirements at each of its research institutes. From a development perspective, however, diversity in the application of single system can introduce a number of challenges. The relatively decentralized organization of the University dictated the need to strike a delicate balance between standardization on the one hand and customization on the other.
In October 2012, a formal publications survey was sent to the newly established focal points which focused on the scope and management of publications across the UNU system. In brief, the survey results reinforced the conclusions of the initial stock-taking exercise that began in April. While there were similarities across the UNU system, there also were a number of institute-specific practices in place. Among the discrepancies that were reported, there were several key variations that would need to be addressed in order to gather and manage content in a unified system.
The first had to do with the extent to which UNU publications were online at all — and in some cases, whether or not the publications were in digital format. As the majority of universities and libraries have experienced in recent years, the move toward digitization of paper records is a massive, resource-intensive undertaking requiring years to complete.
Fortunately, UNU remains relatively young compared to many of its peers, which meant that the majority of the University’s publications existed in soft copy of some form. What was not so consistent, however, was the extent to which robust online repositories had been developed with the purpose of collecting and making available the body of UNU’s research work. There were hard-copy archives and offline databases of records that would need to be accounted for and somehow integrated into the new system.
Even for institutes that had made their publication records available through a web-based system, the variation that existed among these platforms posed another distinct challenge. A notable difference had to do with nomenclature. There were “book chapters” and “book sections” and, at the same time, “policy reports,” “policy briefs” and “policy papers” among similar discrepancies with other publication types. After establishing a global overview of the scope of this variation, it became evident that a consensus would need to be reached in order to present the University’s work in a more organized fashion.
Overall, there were strong indications of support for more-standardized criteria or guidelines in particular areas (one simple example being a more unified set of criteria for works that are counted as UNU publications). By adopting an end-user perspective, it became clear to all involved that better coordination and more consistency — in terminology, definitions and metadata, for example — could help improve access to UNU’s research outputs.
The year 2013 saw considerable progress in building the prototype of the repository. There were, however, a number of technical and organizational challenges along the way.
In broad terms, one of the most significant and time-consuming aspects of the project had to do with metadata. As the technology for institutional repositories has grown, so too has the need for a set of standards in the way that bibliographical information is presented, stored, and attributed. While there is no single solution, there has been substantial progress made toward consolidating the global set of possible metadata types into more consistent, workable sets.
Naturally, for an institution like UNU, a preliminary step in establishing a more unified system for its publications demanded a rigorous and careful analysis of the practices currently in place. The majority of the challenges and achievements in this regard have been documented in detail in the digital.unu.edu post by the project’s technical lead, Conor McTernan. (We invite communications and IT professionals to read about this crucial aspect of the development stage and share your comments and insights.)
With the launch of UNU Collections, a sufficient set of data has been imported and integrated into the repository to allow for at least a preliminary assessment of how this project has met its original goals and expectations. Some of the more significant measures that the project team took to fulfil the aims of this initiative are summarized below.
Removing discrepancies in the way that publication data was classified and described opened the door to defining new ways of interacting with UNU research. UNU Collections enforces international standards for bibliographical metadata and condenses the overall set of publication categories in use across the University system. Pooling together and anchoring works to standard descriptors — e.g., date format, author name order or the displayed name of a research institute — significantly enhanced a user’s ability to find specific content and, at the same time, discover related content.
Where copyright allows, UNU Collections aspires to make the full text of all UNU research available for download (typically in PDF format). By interfacing with the RoMEO database of the University of Nottingham’s Sherpa Services, UNU’s researchers and publications professionals can easily find which works can legally be published in the UNU repository (self-archived), when they can be published (for example, a year after initial publication), and whether or not any restrictions apply to uploading the original publisher’s version. It is anticipated that as more researchers take advantage of this invaluable tool, more research that can be shared through UNU Collections will be discovered, thereby enhancing the availability of important scholarly literature to the individuals and organizations that need it.
Storing UNU’s research in a single location that is easily searchable and sortable is naturally highly useful for those wishing to access findings of a particular research project. At the same time, UNU Collections offers dividends to the University’s communications professionals through an enhanced capacity to quickly and effectively respond to specific requests for research outputs.
With a unified database that is carefully catalogued, it is now becoming possible to solve the technical challenge that I posed earlier. What has UNU published recently on climate change, for example, and who are the University’s leading authorities on this topic? UNU Collections offers the analytical and statistical tools that can make this type of search possible, thereby taking an additional step towards ensuring that the relevant research is provided to appropriate outlets, in an efficient manner and on a global scale.
Leadership. An undertaking at an organization-wide level requires significant time and energy to get off the ground, but along the way it is crucial to maintain momentum. UNU Collections was no exception to this basic rule, and as an in-house initiative, effective leadership could not be undervalued. The working group comprised not only the core project team in Tokyo but a wide network of collaborators across the University system, each with their own busy schedules and assignments unrelated to the development of the repository. There needed to always be someone at the helm, not only steering the direction of the project but also strategically allocating limited resources to meet deadlines and deliver on promises.
Support. Securing sufficient buy-in is a necessary prerequisite to setting a productive trajectory for a project. As most global organizations can attest, what headquarters may consider as an ideal solution to an organization-wide issue may be viewed from a different perspective by local offices and departments. Given the diversity of practices across the University, there was concern from the beginning that such a large-scale initiative might take years to get off the ground. And while the initial approach taken in this project was, to a certain extent, “build first and ask questions later”, it was nonetheless crucial to develop a consensus in the early stages and to incorporate as much feedback as possible.
Flexibility. Good ideas for a project sometimes do not see the light of day, while even initiatives that do get off the ground risk falling short of reaching their long-term objectives. Sustainability was at the heart of every crossroad that the project encountered. Once the repository had been populated with a critical mass of data (which in hindsight seemed to be the easy part), the system needed to be extended across the global UNU community in a way that would ensure its continued use. One example of the type of flexibility required in this regard concerns the workflows behind publishing content in the repository. Where one community may function more efficiently with a channeled model of uploading — i.e., publications flowing through and approved by a dedicated reporting line — other communities with fewer personnel may operate more efficiently with a distributed model one where single authors can upload their own work.
Looking toward the future, it is important to recognize that the development of a new platform like UNU Collections is only one stage in the larger picture of institution-wide solution deployment. Implementation, maintenance and improvement based on user feedback are the longer term (and some may argue, more challenging) aspects of ensuring system integrity, sustainability and, ultimately, impact.
With the launch of the repository, it has been instructive to reflect on the challenges and successes encountered thus far, though it also is equally important to bear in mind the work that lies ahead and anticipate the potential obstacles that would need to be overcome. Moving forward, the task of the global working group turns toward not only continuing efforts to import past content, but also of ensuring the accuracy and quality of how the University presents its forthcoming research to the public. Improvements to the administrator interface will ideally serve to minimize reporting redundancies and excess data entry through the development of streamlined entry and approval workflows.
As with any major initiative seeking to bridge the inevitable gaps that, over the years creep, into large, decentralized organizations, UNU Collections will never be a perfect solution for bringing UNU research to the web. Perfection, however, is forever the enemy of progress. It is hoped that UNU Collections will continue to grow, and that the efforts over the past years by the many people involved in this initiative will contribute to fulfilling its original mandate: to further enhance the impact of the important work being done by researchers across the global UNU system through providing a robust, unified, and user-friendly web-based platform for accessing the University’s publications.