Abstract
The United Kingdom Serials Group (UKSG) is actively involved in key industry initiatives to study issues affecting the scholarly information chain. The Transfer project, Knowledge Bases and Related Tools (KBART), Journal Usage Factor, and the Publisher and Institutional Repository Usage Statistics (PIRUS) project complement one another by wrestling with related aspects of core metadata issues. Problems with metadata transmission and use affect how librarians provide access to and maximize return on investment for resources that are purchased, collected, and archived. How those problems are addressed has a direct effect on how scholars and users access, and qualitatively and quantitatively assess the value of online resources.
INTRODUCTION
Ross MacIntyre is currently employed within Manchester Information and Associated Services (Mimas), the U.K. National Data Centre at the University of Manchester. He is the service manager for three of the largest U.K. information players (Web of Knowledge Services for U.K. Education, U.K. PubMed Central, and Zetoc) and is also responsible for digital library research activities. Because of his professional responsibilities managing systems dependent on data, he has a strong interest in supporting the development of metadata standards. In addition, his role as both an elected United Kingdom Serials Group (UKSG) committee member and an employee of a UKSG organizational member places him in an ideal position to summarize and discuss the current status of specific projects in which he is involved through UKSG: Transfer, Knowledge Bases and Related Tools (KBART), Journal Usage Factors, and Publisher and Institutional Repository Usage Statistics (PIRUS).
UKSG TRANSFER CODE OF PRACTICE
The Transfer project was one result of a 2006 study commissioned by UKSG to uncover sources of inaccurate metadata and to explore how standards might be used to improve the flow of metadata. As part of that study, EBSCO disclosed that there were 5,121 unique titles (16 percent of their title file) exchanged between publishers from January to October 2006. This meant that EBSCO had to make more than 47,000 changes to their title file.Footnote 1 The enormous expenditure of time and maintenance to adjust the metadata to ensure customer access to subscribed content strongly demonstrated the critical lack of industry standards for metadata formats.
The aim of the project is to address the frequent problems with seamless access encountered by users when titles move from one publisher to another. Hypothetically, these problems will be significantly diminished by establishing a set of universally adopted industry standards and guidelines to be utilized whenever titles are transferred from one publisher to another.
The Transfer Working Group, chaired by Ed Pentz of CrossRef, consists of representatives from several principal stakeholder groups, including the scholarly publishing, intermediary, and library communities.Footnote 2 There are four mutually agreed-on goals: (1) to ensure that journal content remains easily accessible by librarians and readers when there is a journal transfer; (2) to ensure that the transfer process occurs with minimum disruption; (3) to establish explicit obligations for transferring publishers and receiving publishers; and (4) to establish best practices to help publishers be more efficient.
The project participants understand that effective collaboration will be mutually beneficial. The Transfer Code of Practice Version 1.0, released in April 2007, was positively received within the community. The group continued to refine the code and with the release of Version 2.0 in September 2008, publishers were invited to endorse and adhere to the code. To date thirty-one publishers have endorsed the code.
When a publisher agrees to honor the code, it means the publisher is willing to meet an explicit set of obligations. The transferring publisher will agree to the following:
| • | To honor perpetual access rights to ensure continued access where rights have been granted (either the transferring publisher or receiving publisher can agree to provide access) | ||||
| • | To maintain access during the transfer, and maintain content online up to six months if receiving publisher is not immediately ready | ||||
| • | To make digital content files available within four weeks of signature of contract or four months before effective transfer date, whichever is later | ||||
| • | To notify subscription lists within the same timing of digital content availability | ||||
| • | To communicate details of the subscription data to consortial subscribers with perpetual access and lapsed subscribers | ||||
| • | To provide journal URL redirects for twelve months or place a link to the receiving publisher | ||||
| • | To communicate the transfer as soon as possible after the contract is signed or no less than two months before effective transfer date | ||||
| • | To follow CrossRef Digital Object Identifier (DOI) name ownership procedures | ||||
The receiving publisher agrees to a matching set of obligations:
| • | To honor perpetual access rights granted by the transferring publisher | ||||
| • | To provide access from effective transfer date but must permit transferring publisher to keep content online if not ready for full transition | ||||
| • | To maintain content in archives (receiving publishers are encouraged to continue existing archival arrangements) | ||||
| • | To communicate transfer specifics as soon as possible after contract or no less than two months before effective transfer date | ||||
| • | To contact all existing subscribers when lists come from transferring publisher | ||||
| • | To follow CrossRef DOI name ownership procedures | ||||
The UKSG Transfer Working Group will continue working toward universal adoption of the Code of Practice. Their mandate is to ensure that the code fairly protects the interests of all stakeholders in the metadata supply chain. They will provide a forum for complaints about non-compliance, periodically review the code and consider revisions, and develop case studies to demonstrate the impact of the code. Because it is critically important that all publishers be willing to commit to the Code of Practice, librarians, authors, and vendors must continually work to educate publishers about why it matters and to whom it matters most, the users. Librarians should subscribe to the Transfer Alerting Service to inform their advocacy efforts. The success of the group's efforts depends on all stakeholders in the scholarly journal community advocating for the code and actively providing feedback to the working group, particularly when problems are identified. The ultimate success of this project will be measured by broad acceptance and adherence to the code.
KBART: KNOWLEDGE BASES AND RELATED TOOLS
Knowledge Bases and Related Tools (KBART) is a collaborative initiative between UKSG and the National Information Standards Organization (NISO); it began in January 2008. The KBART Working Group was a response to the 2007 publication of the UKSG research report titled Link Resolvers and the Serials Supply Chain.Footnote 3 The report outlined critical problems concerning usage of metadata, OpenURL specifications, and link resolvers. The goals of the working group were to improve navigation of the electronic resource supply chain and to ensure timely transfer of accurate metadata to link resolver knowledgebase developers.
Library users depend on link resolvers to access online library resources. Knowledgebases store the metadata that are used to populate the link resolvers. The knowledgebase data quality depends on the quality of the data that are supplied by the content providers (including publishers and aggregators) to the developers. In the absence of standardized formats and practices, errors become common and can prevent users from successfully connecting to online resources.
Approved January 2010, the NISO/UKSG KBART Recommended Practice is a set of best practices to inform content providers as they are formatting and distributing title lists.Footnote 4 Vendors and knowledgebase developers report that when distributors adhere to these practices, it is much easier to make the small adjustments to the data that allow seamless access to content for their clients and users.
The KBART Phase II Working Group has recently been formulated with new chairs and members.Footnote 5 The overall mission of KBART Phase II is to develop a second Recommended Practice to build on the guidelines for data transfer in Phase I, to educate content providers and knowledgebase developers about best practices and the value of adhering to those practices, and to serve as an information hub for reference and educational materials.
JOURNAL USAGE FACTOR PROJECT
In May 2007, the results of a study published by UKSG in association with Counting Online Usage of NeTworked Electronic Resources (COUNTER) were released,Footnote 6 exploring how online journal usage statistics might “form the basis of a new metric of journal quality.”Footnote 7 The study was conducted in two phases. Authors, editors, publishers, and librarians were interviewed by telephone in Phase I. Phase II was a much larger Web-based survey of authors and librarians. The large set of detailed responses to both surveys indicated willingness from all parties to develop a meaningful Journal Usage Factor and to proceed with implementing such a metric. The conclusions and recommendations for further development are outlined in the study.
The usage factor is defined as an equation: total usage during period x of articles published during period y divided by the total articles published during period y. The development of a meaningful Journal Usage Factor depends on overcoming problems with defining and interpreting the data. The principal problems to be addressed include formulating a consistent definitions for the numerator and denominator of the equation, article usage per year, and article publication date. Another important issue is logically addressing the varying usage patterns by subject/discipline.
Stage two of the project is underway, developing a program of data modeling and analysis using real usage data from a number of content providers, with the aim of identifying potential candidate metrics for longer-term, larger-scale testing. This work is due to be completed in late summer 2010.
PIRUS: PUBLISHER AND INSTITUTIONAL REPOSITORY USAGE STATISTICS
The PIRUS project was funded by JISC. Online usage has evolved to become an acceptable measure of journal value due largely to the development of the COUNTER code. The PIRUS project was driven by the increasing interest in usage statistics derived from the article level and recognition that these statistics potentially measure value. However, the existing COUNTER code does not currently cover usage below the journal level. The PIRUS project sought to study the mechanics of generating COUNTER-compliant statistics and usage reports at the individual article level. The ultimate goal was to enable any hosting entity, such as a publisher, aggregator, or repository, to gather and provide for harvesting statistics about article-level usage according to common standards. The project report was released in January of 2009.Footnote 8
The project was largely successful in meeting the aim. Outputs from the project as specified on the project page were as follows:
| a. | A proof of concept COUNTER-compliant XML prototype for an individual article usage report, Article Report 1: Number of successful full-text article downloads that can be used by both repositories and publishers. In principle, this report could be provided for individual authors and for institutions. In practice, the individual author reports are much easier to generate and are a realistic short-term objective, while the reports for institutions and other entities, such as funding agencies, are more complex and should be regarded as a long-term objective. | ||||
| b. | A tracker code, to be implemented by repositories, that sends a message either to an external party that is responsible for creating and consolidating the usage statistics and for forwarding them to the relevant publisher for consolidation or to the local repository server. | ||||
| c. | A range of scenarios for the creation, recording, and consolidation of individual article usage statistics that will cover the majority of current repository installations. Each repository may select the scenario that corresponds to their technology and implementation. | ||||
| d. | Specifying criteria for a central facility that will create the usage statistics where required (for some categories of repository) and collect and consolidate the usage statistics for others.Footnote 9 | ||||
The final recommendations from PIRUS acknowledge that further research and development will be required to implement an article-level COUNTER standard and protocol. There are challenges that must be addressed to proceed successfully. Proposed protocols and tracker codes must be demonstrated to be extensible by testing with large data sets. The codes must also be able to work with non-article data in major repository environments. Organizations identified as potential clearinghouses must be tested to ensure they can receive, store, and process the relevant metadata and generate usage statistics, and to have their missions and operational specifics conceptualized. The cost of generating usage reports for publishers and repositories must be assessed, and it must be determined how those costs will be allocated among stakeholders. Repositories, publishers, and authors must support the project goals. Intellectual property and privacy rights must also be addressed. The recommendations are being taken forward in PIRUS2 (again funded by JISC) currently underway and due to be completed by the end of December 2010.
SUMMARY
These industry initiatives have a common goal: to address problems that impact the entire scholarly information publishing chain from producer to vendor to librarian to user. The Transfer project promotes the development and adoption of best practices relating to journal title moves. The KBART initiative promotes data element standards for sharing metadata and populating knowledgebases. Journal Usage Factor works toward standardized ways to capture the usage data that can inform analysis of journal value. The PIRUS project moves the concept of measuring value based on usage statistics gathered according to technical standards past the journal level to the article level. The impacts of all four projects depend on continued research, promotion, and cooperation within and among all stakeholders.
Notes
1. Jill Taylor-Roe and Nancy Buckley, “Transfer Update” (paper presented at the Charleston Conference, Charleston, SC, November 9–11, 2006), http://www.uksg.org/sites/uksg.org/files/papers_updatecharleston.pdf (accessed July 1, 2010).
2. UKSG, “Transfer Working Group,” http://www.uksg.org/transfer/people (accessed December 3, 2010).
3. James Culling, Link Resolvers and the Serials Supply Chain: Final Report for UKSG (Oxford, UK: Scholarly Information Strategies, Oxford Centre for Innovation, 2007), http://www.uksg.org/sites/uksg.org/files/uksg_link_resolvers_final_report.pdf (accessed June 29, 2010).
4. NISO/UKSG KBART Working Group, KBART: Knowledge Bases and Related Tools (Baltimore, MD: National Information Standards Organization, 2010), http://www.niso.org/publications/rp/RP-2010-09.pdf (accessed December 3, 2010).
5. UKSG, “KBART Working Group Members,” http://www.uksg.org/kbart/members (accessed June 29, 2010).
6. Peter T. Shepherd, Final Report on the Investigation into the Feasibility of Developing and Implementing Journal Usage Factors (n.p.: United Kingdom Serials Group, 2007), http://www.uksg.org/sites/uksg.org/files/FinalReportUsageFactorProject.pdf (accessed December 3, 2010).
7. UKSG, “Usage-Based Measurements of Journal Quality Research Project Enters its Second Stage,” http://www.uksg.org/usagefactors (accessed December 3, 2010).
8. Tim Brody and others, Developing a Global Standard to Enable the Recording, Reporting and Consolidation of Online Usage Statistics for Individual Journal Articles Hosted by Institutional Repositories, Publishers and Other Entities (n.p.: JISC, 2009), http://www.jisc.ac.uk/media/documents/programmes/pals3/pirus_finalreport.pdf (accessed June 29, 2010).
9. Ibid., 3.