Proposal Language
Project Justification
A nationwide collaboration of organizations is seeking a $598,036 grant for a 36-month implementation effort to develop the Collaborative Collections Lifecycle Project (CCLP). The partnership is led by National Information Standards Organization (NISO), Partnership for Academic Library Collaboration & Innovation (PALCI), and Lehigh University Libraries, with contributions from Big Ten Academic Alliance (BTAA), Canadian Research Knowledge Network (CRKN), Greater Western Library Alliance (GWLA), Center for Research Libraries, Colorado Alliance of Research Libraries, Minitex, Orbis Cascade Alliance, the Boston Library Consortium, the Eastern Academic Scholars’ Trust (EAST), Washington Research Library Consortium (WRLC), Columbia University Libraries, Cornell University Library, Johns Hopkins University Libraries, New York University Libraries, Rutgers University Libraries, Tulane University Libraries, University of Delaware Library Museums and Press, University of Denver Libraries, University of Pittsburgh Library System, Washington & Jefferson College, the Duke University Press, JSTOR, Project MUSE, Ithaka S+R, Index Data LLC, ISSN International Centre, Paratext LLC.
The CCLP will create a suite of best practices and improved standards and develop prototype middleware infrastructure for the development and management of collaborative collections. This infrastructure will support varied implementation models, data interoperability and exchange, and sharing of expertise across a range of institutions and consortia. This project aligns with several IMLS goals but it is most specifically targeted at building greater community collaboration (Objective 2.2). The project supports other IMLS objectives as well such as supporting both collections management (3.1) and promoting access to collections (3.2) by focusing on improving institutional efficiency through partnership to maximize public access across institutions, while also reducing duplication in collection acquisitions and management.
Statement of Project Need
Background - Throughout 2020 and 2021, a large group of more than 25 representatives from the partners named in this proposal, including libraries, consortia, publishers, and technology providers, began meeting regularly to discuss the challenges of and barriers to the laborious process of building and managing collaborative collections initiatives in our respective organizations. These discussions considered the successes and failures of previous efforts, which centered primarily on the inability to effectively exchange data about our collections and collections-related expertise at scale, exposing a shared need and revealing a community of stakeholders eager to focus on reducing barriers to implementing collaborative collections.
Context for Collaborative Collections - Networks of libraries have a long tradition of working together to expand their resources and provide more comprehensive coverage across all subjects through sharing of resources. A variety of models have been developed over the years including consortial purchasing, regional centers of excellence, and even shared collections and services among some libraries, such as the 2CUL initiative. More recently, larger networks of institutions have explored wider adoption of cooperative collections management, which this project defines as a process by which networks of institutions work collaboratively to acquire, manage, circulate, and preserve collections across the network. This may include shared infrastructure for patron discovery, fulfillment, analytics, and models for collaboration and organizational decision making. Networks such as the Big Ten Academic Alliance (BTAA), Center for Research Libraries (CRL), HathiTrust, Ivies Plus Confederation, the Boston Library Consortium (BLC), the Eastern Academic Scholars' Trust (EAST), MetaArchive, and Research Collections and Preservation Consortium (ReCAP), have piloted variations on this approach, each with varying levels of success. This project seeks to overcome serious barriers to wider implementations, several of which have been observed, including the lack of available vendor-neutral interoperable systems, data exchange standards, adequate governance and decision-making frameworks, and assessment tools.
To draw an analogy, consider public library systems with multiple branches. For years, these local library systems have purchased items with an eye toward efficiently building a collection that suits the needs of the entire community, without each branch building its own individual collection, since materials can easily be moved from one branch to another. Patrons can withdraw or return items to any branch, which improves the patron experience. All of the branches work on an interoperable backend system, which allows patrons to know each branch’s item availability, and enables decision making about new acquisitions to be made at the network level. This model works because there is a shared understanding of usage patterns at both the network and local levels, as well as interoperable systems–usually a single system–for cataloging, discovery, and circulation. Similarly, research libraries have developed extensive resources sharing mechanisms and attempt to widen their reach, but have yet to figure out how to build and maintain effective network level collections at scale. The CCLP project envisions a new research library ecosystem where networks of institutions and experts can collaborate across systems to more effectively serve their just-in-case and just-in-time patrons’ needs, while avoiding duplication of effort in their acquisitions and deployment of resources. While this is theoretically similar to the way the branches of a public library system already work together, CCLP will support multiple institutions and consortia using a variety of vendors and systems. We can do this by building trust through development of interoperable data and infrastructure, and best practices for governance and decision-making across a range of participating institutions and organizations.
Increasing Equitable Access and Improving Stewardship - As research information grows exponentially every year, libraries nationally and internationally wish to collaborate in a multitude of ways to improve equitable access to library collections. Through collaboration, institutions can effectively cover core-collection areas and redirect scarce resources to improve the diversity and representation of collections by reinvesting resources in areas of social and cultural under-representation. By working together with trusted partners and in an interoperable, community-owned infrastructure, libraries will be able to reduce unnecessary duplication, increase availability of and access to a wider range of sources, better preserve and maintain these resources, and further enhance new and existing collaborative initiatives while maintaining their autonomy. Smaller publishers, open access publishers, and those that publish under-represented voices will be empowered through increased market share and preferred discoverability in the shared CCLP infrastructure.
Evidence of Need - In a 2020 survey by Levenson and Hess of library staff (n=83) describing the potential benefits of cooperative collections development (CCD), 69% responded that cost savings would be the greatest benefit. Sixty percent noted "increased breadth and depth from access to shared collections”, while 47% thought cooperative collections development would lessen the burden of price negotiations on individual libraries. Similarly 45% thought cooperative collections development would lessen the licensing burden. Forty-two percent thought this approach would reduce unnecessary duplication among libraries. The survey also collected data on librarians’ (n=70) perceptions of the greatest drawbacks they perceived to this approach. Respondents overwhelmingly (93%) selected the complexity of managing cooperative collections development as the biggest barrier. Individuals also referenced vendor resistance (53%) and decreasing autonomy in resource selection (44%) as other barriers. Clearly, to be successful, cooperative collections development needs to be simplified and standardized to become widely adopted. When asked whether the potential benefits of collaborative collections development outweigh its potential drawbacks, 64% responded that they either agreed or strongly agreed, with an additional 26% somewhat agreeing to the statement. The study’s authors concluded, "These types of initiatives require a much higher level of effort and coordination, and this complexity may directly relate to institutions' or librarians' hesitancy to engage in such work. In the same vein, respondents selected the complexity associated with managing CCD activity to be the greatest drawback to success." From its inception, CCLP is meant to be built and governed in deep collaboration between publishers, providers, consortia, individual libraries, and standards communities to take into account heterogeneous perspectives and data workflows. It aims to lower the cross-industry barriers to collaboration by reducing complexity and increasing the trust among potential collaborators. We envision that a healthier collaborative infrastructure will increase overall market share by supporting new collective efforts and new business directions. Reduced barriers should also increase competition and extensive utilization of industry standards will increase overall market effectiveness.
A Failing Marketplace for Scaling Efficient Collaborative Collections Activities - At this stage, there is no infrastructure existing to support collective collections at scale as envisioned by CCLP. Selectors and collection support teams work within the limitations of their own local systems and are restricted to vendor-supplied information (hard to find at best) regarding potential duplication and usage inside their own consortia networks (often inside their own branch libraries). The data ecosystem for identifying available content, its providers/distributors/sellers, and the details about processing an order (such as price, rights, payments, consortial relationships, etc.), are complex and heterogeneous. Consortia and libraries have attempted to circumvent such dependencies by agreeing on collaborative collection building via joint purchasing, shared approval arrangements, or shared selectors, but often such higher-level agreements fail over time, when local collection priorities change and personnel shift. In addition, and with the lack of adequate infrastructure and interoperable data standards, selectors need to constantly be aware of potential duplication across such arrangements, and context-shift between multiple vendor systems dealing with particular collection coverages. The lack of an accepted trusted registry and best practices related to collective and local decisions, standardized and machine readable approval plans and collective MOU’s, lead to a growing gap between strategic prospective collection building and retrospective collection analysis and ultimately to a waste of institutional resources, and collective effort. Libraries and consortia lack a timely mechanism to translate proven longitudinal collection behaviors into operational prospective realities, in support of communal targeted investments. A number of organizations have undertaken a range of projects which will inform this project’s efforts. Among examples of related work are: Exploration of aggregation of holdings information, comparison tools (Gold Rush), and aggregation of usage data (projects GreenGlass and Unsub) and aggregation of ONIX data for potential resource acquisition and deposit. While work has been moderately successful and seen localized implementation, these systems have yet to break through and find widespread adoption as such tools remain separate from the daily acquisitions and selection decisions which also trigger collection lifecycle processes (and the expensive invisible economics related to them). An additional result of such limitations—while libraries for more than two decades have promoted open data, open access, and heterogeneous access to diverse resources in their advocacy efforts—local resources related to the metadata creation needed to allow such content to be discovered have diminished or been outsourced to support local buying behaviors. CCLP, beyond other benefits, can allow a more strategic reallocation of metadata expertise to support much needed growth in collective scope related to open science.
In order to achieve the potential envisioned by cooperative collections development, workflows need to be reimagined if they are to function across multiple institutions, and need to be supported by technology tools that facilitate communication and individual and collective decision making. In addition, decisions that are made in specific, time-bound, collaborative arrangements need to be recorded and analyzed to serve new relevant mission-driven directions and future collection decisions. Beyond these systems, governance and frameworks for collective decision making need to be developed and tested to ensure that they are suited for deployment across a range of institutions and consortia and the experts working within these structures.
CCLP Addressing the Need - The CCLP infrastructure promises to optimize daily, network-first collaboration between libraries on the institutional, consortial, and inter-consortial levels. Its availability will improve equitable access to library acquisitions by giving small publishers and open access providers the preferred logistical footing as established for-profit publishers, and will encourage healthier scholarly communication lifecycle activities by directly partnering with university presses and not-for-profit providers. Dashboard insight into local and network level collections, their usage, and preservation status will assist heads of library collections and individual selectors in collaborating while increasing data-driven decision-making and coordination of prospective collecting to emphasize what is of unique value to their communities. The common CCLP infrastructure will also allow for greater logistical efficiency, with centers of excellence acting as functional designate nodes to enable a more sustainable overall ecosystem. For example, selectors in certain disciplines across multiple organizations will be able to treat overall collection development workforce and processing expertise as serving one great collection. They will be able to contextualize local decisions such as firm orders, approvals, deaccessioning, and annexation projects, across systems and workflows and within an interoperable, data-informed, and network-first context. CCLP will provide them a shared virtual-meeting point to partner with other selectors, to communicate, coordinate, and actively collaborate with each other in order to achieve the goal of more relevant titles accessible to all patrons via resource sharing mechanisms. It will allow them to apply readily available data from individual purchasing decisions and users’ activity to inform new collection development models, new purchasing plans, and to create data-informed MOUs and approval agreements. It will extend functionality that will alert of potential duplication and a means to know what is low use and of low interest before placing an order necessitating processing workflows. Once selected, newly acquired collections can be directed for physical and digital preservation, retention, and archiving, or to receive community centered metadata enrichment. These distinct management actions could be done for certain collections in specific languages, for areas of local significance, or for any other materials of value. Such activities will allow hubs of processing excellence to specialize and grow and increase overall accountability via process and cost awareness. Through visibility, CCLP will enhance transparency and trust and encourage additional network reliance. The community owned vendor-neutral collaborative collections lifecycle platform will flexibly address changing needs over time and retain the primacy of libraries collections' buying power by supporting new and scalable collection development pathways.
Target Audience and Beneficiaries
The infrastructure and trust created by this initiative will benefit patrons of those libraries working in collaboration using CCLP tools. The main target audiences for this project are: 1) The selection teams and leadership within libraries and consortia, including those librarians involved in materials selection and acquisitions, who will benefit from the infrastructure support around their expertise and partnerships, and greater transparency about availability and use of selected resources; 2) The Deans/Directors/AULs leading library organizations/units who will be more effective in managing their institutions, directly through better analytics about collection related expenditures in order to achieve greater return on investment within their organizations; 3) Consortia, either officially organized or informal groups, which will benefit from decreased barriers to collaboration and infrastructure to sustain existing and innovative partnerships.
More broadly, there will be many beneficiaries from this initiative. They include:
Library Users/Patrons, who will benefit directly from the collections and services libraries provide
Describers, library, publisher, and systems staff involved in materials description, who will benefit from a common system for managing and discovering metadata
Content providers of any sort, who will benefit from easier and timely access to library systems, relevant data, and decision makers across institutions
Suppliers of library systems, who will benefit from increased interoperability and logistical effectiveness
We see this work offering direct, short-term benefits for libraries and patrons. Library patrons will benefit from a greater breadth of collections access that they can draw upon. The CCLP infrastructure will also allow the focus on expertise and resources needed to mitigate the risk involved in more ambitious strategic collective action because it is shared across many partners. This model could also be adapted by non-academic libraries seeking to develop collaborative collections development strategies in their own networks.
Project Work Plan:
The first phase of this effort will be the development of a community-based governance structure (Deliverable #1) and detailed project roadmap with requirements, specifications, and feature prioritization mechanisms (Deliverable #2). A simultaneous data gathering initiative managed by Ithaka S+R will provide a detailed assessment and documentation of the landscape and classify existing standards and current practices of organizations engaged in collaborative collections projects, including those pilots launched by project participants. This assessment will also include interviews of key community leaders regarding their organizational requirements and expectations of a successful outcome (Deliverable #3). Building on these elements, a community working group will then develop model workflows, model user experience, and activity paths based on defined personas engaged in CCD at different management levels in libraries, consortia, and publishing (Deliverable #4). Based on these components, the group will then build mockups and wireframes of key components of the needed infrastructure (Deliverable #5), taking into account existing systems and collaboration workflows and policies, combined with the identified gaps in those systems based on the goals of this project. The team will also model a community-based implementation structure describing interactions, specifications, and feature prioritization mechanisms (Deliverable #6). The team will develop prototype middleware tools, where those tools do not exist, in partnership with technology vendors. Where existing tools may exist but may need customization or iterative improvement, this team will prototype those adjustments (Deliverable #7).
The project will be managed by Todd Carpenter, Executive Director of NISO, Jill Morris, Executive Director of PALCI, and Boaz Nadav Manes, University Librarian at Lehigh University. These three will oversee the grant administration. The team will be supported by a Project Manager to be hired at NISO and a variety of staff contributions from PALCI, Lehigh, and the other partners in the project. The project will also be supported by a Steering Committee composed of volunteers who will oversee organizational contributions to the project. This governance committee will guide the overall project’s goals and provide project management oversight. It will also manage various working groups undertaking elements of the projects as it progresses.
The project will pilot and explore interoperable, open source middleware with modular applications that will be closely aligned with current collective collections visions and practices. The middleware applications will be developed based on an open standards architecture and will support the flow of data about disparate library collections. This will include holdings information, contractual information, retention obligations, usage data, aggregation of library staff and subject matter expertise, local/consortial/group-based insights, and publisher/marketplace information necessary to support collaborative decisions at both the local and cross-institutional levels. Initial planned applications may include: A) An aggregated shared index and knowledge base in which libraries/publishers can share data about their collections and expertise; B) A discovery mechanism for library staff to support searching and browsing for content, information, and human resources; C) A communication application that will support interactions across institutions; D) Data aggregation, visualization, and reporting; E) Negotiation and group purchasing decision support protocols.
After a round of public comment, the combined model and toolset will be vetted by NISO standards committee leadership and, if approved, published openly as a NISO Recommended Practice. All of the other deliverables for this project will be made freely available using Creative Commons Attribution 4.0 (CC-BY) license or using an MIT open source license, as appropriate. Training and outreach materials will also be developed.
This effort will proceed in two phases with an anticipated third future phase planned to support full implementation following this project’s completion, some of which will overlap during the time frame outlined below. Some elements of the phases will occur simultaneously as work on the previous phase reaches its final stage. The project will be timed such that core elements of later stages will occur in a timely manner, so as to allow early stage work in the next phase to begin where contingencies are critical to determining its direction. Project management structures will be in place to ensure adherence with the proposed timeline. Prior to the receipt of the grant, the project team will advance a NISO New Work Item Proposal, to gain the necessary approvals to launch the initiative as a NISO standards activity, contingent on the project's funding.
Phase 1: Requirements gathering, governance establishment and roadmap definition
Phase 1 of this project is focused on the development of the social infrastructure necessary for the subsequent creation and adoption of technical infrastructure. The team has already discussed many of these topics, but consensus development around the group’s initial ideas will be critical to getting the necessary buy-in to these ideas to ensure that they are broadly adopted by the community. Phase 1 outputs will provide this foundation for the subsequent development work. Once the project is funded, a Steering Committee will be formed from the existing stakeholders. A public call will also be issued to solicit additional community participation, as well as on the other working groups that will be formed. Committee participation will be vetted in collaboration with NISO’s Information Discovery and Interchange (IDI) Topic Committee. An essential element is the creation of a governance structure for the project including the participation of existing stakeholders and other prospective stakeholders who may respond to a public call for participation on the Governance Working Group, consisting of approximately 10-15 volunteers. The group’s first task will be to agree on the community governance structure for decision making within the project. Ithaka S+R will facilitate these discussions, serving in a consultative role, and provide advice guiding it to a successful outcome.
Ithaka S+R will conduct a landscape review (i.e., a detailed assessment and documentation of the landscape) of organizations engaged in collaborative collections projects, including those pilots launched by project participants. This work will be conducted through a combination of desk research and interviews. As part of this landscape review, Ithaka S+R will classify existing standards and current practices. This landscape review will be published as an Ithaka S+R report or issue brief and will constitute Deliverable 1. Ithaka S+R will separately conduct a needs assessment regarding organizational requirements and expectations of a successful CCLP outcome. This needs assessment will be based primarily on interviews with key community leaders, potentially supplemented with focus groups and other forms of organizational and consortial engagement. When completed, it will be reviewed by the Governance Working Group and is Deliverable 2. A second set of working groups will then develop model workflows, model user experience and activity paths based on defined personas engaged in collaborative collections development at different management levels in libraries (Deliverable 3). This work will be undertaken by a committee of approximately 15-20 volunteers working for roughly three months. The group will break into subgroups to develop detailed workflows for the following personas: Consortia, Publishers, Standards, Metadata, Acquisitions, Selection, Providers, Deans.
Based on Deliverables 2 and 3, a subsequent working group will then detail a functional roadmap of key components of the needed infrastructure (Deliverable 4), taking into account existing systems, collaboration workflows, and policies identified in those deliverables. This roadmap will include guidance on technological tools to be developed or adapted into existing systems, along with policy guidance on community best practice for implementing cooperative collections development within libraries. After a round of public input, the combined model with its governance components will be vetted by NISO standards committee leadership and, if approved, published openly as a NISO Recommended Practice. This recommendation will form the basis for the technical development of the middleware systems.
Phase 2: Development team organization, initial application prototyping, public advocacy
The second phase of this project will focus on advocacy, partner engagement, and the development of prototype systems that can implement the technological tool guidance provided by Deliverable 4. This phase will begin with the establishment of a prototype development team, including one paid developer and a user experience specialist supported by this grant, and additional in-kind contributors with applicable expertise. They will identify necessary systems development resources, programming resources, and various in-kind contributions from partners in this project. The organization of Phase 2 will begin before all the Phase 1 outputs are fully completed to allow iterative-design-thinking development. The prototype development team will establish a structure of virtual operation to ensure project timelines and development goals are met.
The prototype development management and team members will determine the scope of critical functional areas of the necessary systems components. It is possible that not all of the modules described here will be necessary, depending on the outcomes of Phase 1 but it is anticipated that these four key applications will be considered as needed, such as a global product catalog (to understand the universe of available content), a collections analysis tool (a unified view of actual holdings), a usage analysis tool (as well as decision-making), and decision makers’ interface (a tool to purchase and support decision-making and communication). The global product catalog will be a comprehensive resource where libraries can turn to identify options for adding content to their collections. This will be derived from publicly available feeds of data from publishers (normally distributed in ONIX format or via MARC) about new and existing product offerings and will serve as a resource institutions can use to manage their acquisitions activities. The collections analysis tool will provide a timely view of the holdings of the participating institutions, so that decision makers can discern collaboratively which items to collect or withdraw and the number of items that might be needed in the collection. A usage analysis tool will also support the collection analysis tool, through the aggregation of per item usage data from print and online sources. Finally, a decision making interface tool will allow individuals in participating institutions to advance their acquisitions decisions by supporting the collective decision-making process, communication, and ordering and fulfillment tracking of that order. For each of these applications there will need to be functionality to populate the infrastructure with data from its various data stores around the network. These system components will then be mapped to the framework and prototyped using user experience design methodologies to meet the demands of library staff, management, and library patrons.
Additional work on the necessary data exchange between the prototype systems and other library or publishing systems will also be documented and proposed as new standards. Potentially existing standards, such as those for interlibrary loan, circulation, usage data collection, etc., may simply need to be adapted and extended to work in this new environment. Other newer specifications may need to be drafted and consensus reached on those new standards. Phase 2 will include this consensus development work, guided both by the Phase 1 outputs and the experience of prototyping the Phase 2 modules.
Once published, any standard requires a variety of outreach and promotional efforts to socialize the recommendations. Promotion of and advocacy for engagement with the Phase 1 outputs will entail a number of in-person and virtual events, including a mix of both large presentations and small group discussions, as well as virtual presentations and published papers. These will describe different elements of the outputs with the aim of continuing to gather additional input on the elements of the project, as well as to recruit partners in the deployment of tools being developed. All of these public sessions and papers will be freely available.
The components of the system will be prototyped and provided to the community via the MIT open-source license, although it is also envisaged that proprietary software providers will also engage in the project, developing their own tools that will be integrated into their offerings. Part of the Phase 1 and early Phase 2 activities will include outreach to commercial vendors providing library systems to assess their willingness to participate and develop these tools commercially, either through open source development as part of a future phase or separately in their own environments.
Following completion of this work, project organizers anticipate a third future phase to deploy and fully implement the prototype system for full-scale evaluation.
Project Deliverables
The first phase of this effort will be the development of a community-based governance structure (Deliverable #1) and detailed project roadmap with requirements, specifications, and feature prioritization mechanisms (Deliverable #2).
A simultaneous data gathering initiative managed by Ithaka S+R will provide a detailed assessment and documentation of the landscape and classify existing standards and current practices of organizations engaged in collaborative collections projects, including those pilots launched by project participants. This assessment will also include interviews of key community leaders regarding their organizational requirements and expectations of a successful outcome (Deliverable #3).
Building on these elements, a community working group will then develop model workflows, model user experience, and activity paths based on defined personas engaged in collaborative collections development at different management levels in libraries, consortia, and publishing (Deliverable #4).
Based on these components, the group will then build mockups and wireframes of key components of the needed infrastructure (Deliverable #5), taking into account existing systems and collaboration workflows and policies, combined with the identified gaps in those systems based on the goals of this project. The team will also model a community-based implementation structure describing interactions, specifications, and feature prioritization mechanisms (Deliverable #6).
Documentation of the prototypes and deployment of an outreach plan to widely circulate sharing of the project’s results, encourage adoption, and assessment of the project (Deliverable #7).
Diversity Plan
Many libraries are seeking to bring the lens of diversity, equity, and inclusion to expanding their collections in order to include voices and perspectives that have been historically marginalized or excluded altogether. A challenge they collectively face in doing so, however, is that the very technologies and systems they depend on to tackle the monumental task of diversifying collections are written by and for a majority largely responsible for the marginalization and disadvantaging of people of color. Bias is often embedded in the very code, programmed into the logic of our collections systems. Through a community-based, open, and inclusive approach, the project will create a system of automated workflows rooted in principles of equity and justice, programmed from the ground up to be inclusive of diverse perspectives and sensitive to cultural variety. The project will seek out a deep engagement with institutions and participants representing historically under-served populations and will directly engage diverse institutions’ staff in the project leadership and working groups. The project will improve the diversity and representation of the collective collection by facilitating net reinvestment of duplicative resources and expertise into areas of social and cultural under-representation. It will also allow partners to redirect resources towards areas of national or international emergency collecting and preservation.
Project Results
A core element of the success of this project will be in establishing community and networks of trust among institutions. Convening an open Collaborative Collections Lifecycle Community Hub with diverse participation from academic libraries, consortia, publishers, technology organizations, and other library service providers will enable the development of a shared vision, business practices, and infrastructure needs. The development and initial testing of the CCLP middleware prototype will lead to more efficient, sustainable, and responsive library collection-building activities and will encourage further growth of network level partnerships.