To help the Resource Discovery Taskforce members think about a vision for UK HE resource discovery JISC commissioned a report to look at how other nations are addressing resource discovery and what some commercial companies are providing in this area.
The report was written by Rightscom and is a detailed look at the work on resource discovery of:
- The Netherlands
Here is the executive summary of the report to whet your appetite:
This report describes and analyses some examples of resource discovery services (in Australia, Canada, the Netherlands and Sweden) incorporating national union catalogues, and also examines some commercial and non-commercial services which incorporate resource discovery as a key part of their operations.
In partnership with RLUK, JISC has set up a Resource Discovery Taskforce to focus on defining the vision and requirements for the provision of a shared UK infrastructure for resource discovery for libraries, archives and museums to support education and research. It is important for the Taskforce to be aware of relevant services, infrastructures and technologies used in other countries and in the commercial sector for resource discovery. The purpose of this report is to provide the Taskforce with that information. The report should be read in that context.
It is not possible to condense the description of each service into this Summary in a meaningful way. However, there are some common strategic challenges facing the national resource discovery services, and they are taking similar approaches to solving them.Common challenges include:
- Making services more welcoming to those accustomed to using Google and social networking sites
- Enabling discovery across media types
- Ensuring that the information contained in library catalogues is exposed as widely as possible
- Making it as easy as possible to move from finding an item to getting it
Common solutions include:
- Redesigning interfaces
- Incorporating relevance ranking, faceted search and results clustering, as well as features such as ‘Did you mean…?’
- Importing book cover art, reviews and tagging from sources such as Amazon and LibraryThing (these services are examined in the report)
- Improving and extending metadata
- Taking library data out to users in various ways rather than expecting them to come and find it. Examples include exposing records via Google BookSearch, Google Scholar and Yahoo! either directly or via WorldCat; embedding search boxes in Facebook; making the entire catalogue available in Linked Data
- Tying finding more tightly to getting, for example by deep linking to the catalogues of a user’s local library; providing pre-paid non-mediated Inter Library Loan accounts for users; linking to bookshops’ websites; experimenting with home delivery
The national resource discovery services face some dilemmas and barriers as well:
- Very limited resources in comparison with commercial services, especially in the current economic situation. This applies whether the services are directly taxpayer-funded or dependent on subscriptions from libraries
- Balancing the need to integrate into the wider network for greater visibility with the desire to not to become too dependent on large and powerful organisations such as Google
- Breaking down technical and human/organisational barriers in order to present a more coherent discovery environment to users, for example, enabling single search across media types, integrating multiple portals
- Making crucial technology decisions (for example, on open source, buy or build, semantic web technologies) when the environment is changing so rapidly
- Balancing public accountability and a culture of caution against the need to take risks and adopt a ‘beta’ approach to development
- The need to scope what national resource discovery services should and should not encompass e.g. there may be a temptation to try to build a community where it’s not appropriate or feasible
For the sake of convenience, the other services are referred to as ‘commercial’, though their business models differ considerably.
Drivers for these services include:
- Shaping their services to compete effectively, which in turn may or may not imply selling more to the user, but in all cases implies creating and sustaining users’ trust
- This is bound up with creating an environment in which users value and trust the contributions of other users, and satisfying their desire to pass their own knowledge and opinions on to others
- To be competitive, these services must be able to provide user-friendly interfaces, additional features, depth of content and a sufficiently large user community to maintain content and ensure freshness
- A business model that enables partnerships is critical
Challenges and barriers for these services:
- The main barrier facing most of the services is the presence of a strong leading competitor, making it difficult for the others to get traction
- Small numbers of users and small volumes of metadata restrict the usefulness of both automated and human-based recommendation and evaluation services; in addition, a relatively small proportion of active users means the services have to attract large total user numbers
- A key challenge for smaller services is to retain access to (preferably open) sources of data required for sustainability
Conclusions specific to national resource discovery services:
- All of the services examined in this report understand that change is vital if library catalogues are to retain relevance and visibility in the wider networked discovery environment
- Some have made more progress towards change than others, but they are on similar paths, albeit with variations according to circumstances and strategies
- There is a need to make more concerted ongoing efforts to understand users’ needs and behaviours (not only when developing new interfaces) and where appropriate, segment their user bases and market services more effectively to these different groups
- Community features need scale and the services are right to follow the path of relying on tags and reviews from e.g. LibraryThing and Amazon, though this does imply dependence on services which are outwith their control
Overall conclusions, which apply to all services:
- Scale is vital for user-generated material: both the size of the user community and the size of the metadata collection will make a considerable different to a service’s ability to attract and retain users.
- The increasing dependence of resource discovery services (both commercial and non-commercial) on a number of large external datasets is of interest. This interdependence seems to be at the data level: as yet, functionality is not shared between the resource discovery services studied here. Although not at the moment an area for concern, dependence on data from other resource discovery services can cause some problems—as the recent reaction to OCLC’s change of policy on WorldCat records illustrates. In the long term, it may pay to monitor the risks of a monoculture for data. Models that allow many smaller services to contribute data, or for their data to be used in real-time, may offer some insurance against data monocultures.
- More needs to be done to understand user journeys: individual sites monitor what users do, but the way they move between different sites is not known. This would help analyse how users get to resource discovery services and what they do once they have finished. Understanding user journeys would also help to elucidate the relationship between (for example) either identifying a book through Amazon and then using a union catalogue to locate a copy to borrow; or finding something in a union catalogue that was not available to borrow locally and purchasing a copy.