Link Resolvers and the Serials Supply Chain

1. Link Resolvers and the Serials Supply Chain A UKSG-funded research project undertaken by Scholarly Information Strategies Presented by: James Culling, Online Project Manager, Oxford University Press Briefing Session at UKSG Annual Conference, Warwick 2007

2. Publicising the results of this study • UKSG conference briefing sessions • Summary article in Serials – July 2007 • Full report available from UKSG website following article publication

3. Agenda • Industry context • Project context • Methodology • Results • Description of the supply chain • Issues and barriers • Recommendations • A centralised alternative • Summary of findings • Where next?

4. Industry context Link Resolver Service Menu Link-to Syntax OpenURL A&I database Publisher site Article/Chapter Book or Journal Table of Contents Book or Journal Title Homepage Source Target Knowledge Base

5. Industry context

6. Industry context

7. Project context “…just because full-text finding tool vendors update their products regularly does not mean that the lists are actually up-to-date, because full-text finding tool vendors get updates from content providers who have various updating schedules and practices, and thus are of varying quality.” Chen, X. “Assessment of Full-Text Sources Used by Serials Management Systems, OpenURL Link Resolvers, and Imported E-Journal MARC Records.” Online Information Review 28.6 (2004): 428-434. <http://dx.doi.org/10.1108/14684520410570553>

8. Project context “…the quality of the service could vary widely depending on the accuracy and completeness of the SFX KnowledgeBase. The library has a staff member who “knows better than the vendors do what is in their own databases”, noting that San Marcos reports roughly 30 errors per month back to Ex Libris. Therefore, it took more time to maintain the KnowledgeBase than initially thought.” Wakimoto, J.C, Walker, D.S, and Dabbour, K.S. “The Myths and Realities of SFX in Academic Libraries”. The Journal of Academic Librarianship 32.2 (2006): 127-136. <http://dx.doi.org/10.1016/j.acalib.2005.12.008>

9. Project context • Study objectives: • Understand present “distributed” supply chain • Expectations, roles, relationships • Establish what is not working well • Consider how to address issues • Practical steps, best practice, role of standards in accelerating/automating data transfer • Consider alternative “centralised” model • Enthusiasm for it? Who might contribute?

10. Methodology • Stakeholder interviews • Publishers and other content hosts • Link resolver vendors • Librarians • Subscription agents • Others • List serv mailings (targeting librarians) • Online survey of librarians

11. Description of the supply chain Requested by library Publishers/Content Hosts Requested by library Subscription Agents Pull Push Pull Push Internal Library Systems Holdings / subs files Content packages Library Link Resolver Supplier* Supplier Hosted Resolver either or Library Hosted Resolver Master KB * There are a number of resolver suppliers building their own proprietary Knowledge Bases (KBs) for the market

12. Stakeholder Knowledge Base Supply Chain Role(s) Relationship With Stakeholder’s Expectation of Relationship Publisher / Content Host To make collection data description available to Link Resolver Supplier. Link Resolver Supplier That Link Resolver Supplier will process/upload data in a timely fashion and distribute on to the Library. Where relevant, to make subscription files/holdings details available to Library for use in customising the Publisher/Content Host’s knowledge base target to reflect local conditions. Library That library will request/download data as and when it needs it. To provide details of a suitable inbound linking syntax for delivering Library users to the content (Target). Link Resolver Supplier That Link Resolver Supplier will configure its system to utilise the linking syntax for users connecting to the Publisher/Content Host’s material. Where relevant, to provide details of OpenURL support (Source). Link Resolver Supplier and Library That Link Resolver Supplier will check its system integrates satisfactorily and that Library will enable OpenURL linking on the Publisher/Content Host’s web site for its users. Stakeholder Knowledge Base Supply Chain Role(s) Relationship With Stakeholder’s Expectation of Relationship Subscription Agent To make collection data description available to Link Resolver Supplier [for agent gateway product]. Link Resolver Supplier That Link Resolver Supplier will process/upload data in a timely fashion and distribute on to the Library. To make subscription files/holdings details available to Library customer for use in customising knowledge base targets to reflect local conditions. Library That library will request/download data as and when it needs it. To provide details of a suitable inbound linking syntax for delivering Library users to the content (Target). Link Resolver Supplier That Link Resolver Supplier will configure its system to utilise the linking syntax for users connecting to Subscription Agent’s content. To provide details of OpenURL support (Source). Link Resolver Supplier and Library That Link Resolver Supplier will check its system integrates satisfactorily and that Library will enable OpenURL linking on the Subscription Agent’s web site for its users. Stakeholder Knowledge Base Supply Chain Role(s) Relationship With Stakeholder’s Expectation of Relationship Link Resolver Supplier Gathering, normalisation and processing of collection data for loading into master knowledge base. Publisher / Content Host and Subscription Agent Data provided is: accurate/current; consistently laid out from one update to the next; sufficiently descriptive (title identifiers, coverage information); structured (comma or tab-separated file format is typical). Mechanism for data to be automatically sent or collected is in place. Gathering and configuration of inbound linking syntaxes to accompany collection data for Targets. Publisher / Content Host and Subscription Agent That linking syntaxes are: flexible (support linking to different levels); well described; do not rely on internal identifiers; do not change over time. Checking (and possibly documentation) of OpenURL support in Sources. Publisher / Content Host, Subscription Agent and Library Link Resolver Supplier may provide documentation regarding OpenURL support by content sites to the Library as part of its service. Distribution of updated knowledge base data to Library. Library That Library will process the update in its own time (if such a step is necessary) and take any other manual action resulting from changes in the knowledge base contents (for example, configure new targets, make live new titles added to an existing subscribed target). Stakeholder Knowledge Base Supply Chain Role(s) Relationship With Stakeholder’s Expectation of Relationship Library To activate ‘standard’ subscription targets in the knowledge base (for example, aggregator database products). Link Resolver Supplier Library expects complete accuracy in the data for these targets. Does not have time to check. To collect subscription details from internal systems (library catalogue, A-Z list, ERM tool etc) for customising other knowledge base targets to reflect local conditions. Other library systems [Internal] Data should be: accurate; current; structured; easy to collect etc. These are all issues internal to the library and the suppliers of its own software systems. To collect subscription details from Publisher/Content Host for customising the relevant knowledge base target to reflect local conditions. Publisher / Content Host Data is: accurate/current; consistently laid out from one update to the next; sufficiently descriptive (title identifiers, coverage information); structured (comma or tab-separated file format is typical). Mechanism for online collection is in place. To collect subscription details from Subscription Agent for customising some knowledge base targets to reflect local conditions. Subscription Agent Data is: accurate/current; consistently laid out from one update to the next; sufficiently descriptive (title identifiers, coverage information); structured (comma or tab-separated file format is typical). Mechanism for online collection is in place. To receive and apply (or have applied) knowledge base updates from Link Resolver Supplier. Link Resolver Supplier Data is accurate and current. Library expects Link Resolver Supplier to have checked the data it received from the Content Hosts before distributing it further. Updates should be provided to the Library on at least a monthly basis. To process knowledge base updates from Link Resolver Supplier in a timely fashion for Users [if such a process is necessary, only likely if locally hosting the solution]. User The provision of an accurate and reliable service to the end customer. Stakeholder Knowledge Base Supply Chain Role(s) Relationship With Stakeholder’s Expectation of Relationship User To use the OpenURL-based linking service provided by the library. Library User expects accuracy in service menus and links at all times. Description of the supply chain

13. Description of the supply chain • Complexities… • Publisher products not sold through agents • Resolver suppliers outsourcing KB work • Interplay of library systems – data staging posts • Consortia KB and institution KB • Who does the KB localisation? • Can libraries create KB content themselves? • Libraries as suppliers of KB data to resolvers • Interaction with CrossRef • Interaction with Google Scholar

14. Issues and barriers • Lack of awareness • Lack of co-operation • Inaccurate and incomplete data • Content package issues • Journal title changes and transfers • Responsibility for data quality • Lack of data standards • Timing issues • Inbound linking issues • OpenURL issues • Role of the subscription agent • Broadening of knowledge bases

15. Issues and barriers Lack of awareness Lack of co-operation

16. Issues and barriers Inaccurate and incomplete data

17. Issues and barriers Content package issues

18. Issues and barriers Lack of data standards

19. Issues and barriers Summary: OpenURL compliance is not knowledge base compliance

20. Recommendations • “COUNTER” for knowledge base and OpenURL compliance

21. Recommendations • Areas for a code of practice: • Knowledge base compliance for publishers and content hosts • Knowledge base compliance for subscription agents • Knowledge base compliance for link resolver suppliers • OpenURL compliance for content providers

22. Recommendations 2. Accelerating and automating knowledge base data transfer

23. Recommendations • ONIX for Serials Online Holdings (SOH) • Web services and ONIX SOH • SUSHI equivalent? • There is a need for experimentation • Content providers  Link resolvers • Subscription agents  Link resolvers

24. A centralised alternative • Appeals to many content providers and librarians • But quality assurance task is a significant barrier to entry • Costs are not insignificant – investment from where? • And competing with existing players = Does not seem viable

25. Summary of findings • Complex/dependent nature of present supply chain • Key issues: • Lack of understanding and co-operation • Lack of clarity and transparency over requirements, including data formats/structures • Knowledge base compliance overlooked as an issue in the industry • A code of practice (similar to COUNTER) would help address problems in the short term • Experimentation with XML feeds needed for accelerating and automating data exchanges

26. Where next? • What is UKSG doing with the research results? • Next steps will be finalised at UKSG board meeting in May • When and where will findings be published? • Summary article in July issue of Serials • Full report available on UKSG website thereafter

27. Thank you James Culling Online Project Manager Oxford University Press Tel: 01865 354851 Email: james.culling@oup.com