This case study makes use of the following, which are described in more detail in the linked resources:
This case study was done as a stand alone exercise by Owen Stephens with input from David Kay.
1. The Problem
The problem statement for this case study is taken in its entirety from the Knowledge Base Plus Business Solution Documentation.
A major issue for electronic and subscribed resources is that the offerings of resources are complex and fragmented, in terms of both processes and data elements (e.g. price, licensing, post-cancellation access).
Collecting and managing the relevant data is costly. Whilst many aspects of the information are common across the whole UK HEI community, currently this information is collected many times over and updates or corrections to the data discovered by one institution cannot be easily shared across the community.
The complex nature of the offerings leads to a number of issues that result in both direct and indirect costs to the subscribing institutions. Key issues include:
- Decision making (such as which resources to cancel or renew) is hampered by the lack of data, and because relevant data is recorded across multiple systems (e.g. quantitative and qualitative measures of the value of a resource to the institution).
- Errors in data can result in institutions losing access to resources, directly impacting on researchers, teachers and students
- Changes to data are often known at a point distant in time from the required action (e.g. a cancellation decision may be made several months before the cancellation is enacted and relevant systems updated). This increases the risk of error.
- Financial transactions related to offerings can be complex. This can be due to local budget governance, the need to record financial transactions in multiple systems, or simply due to the nature of the materials being purchased.
a) Data driven systems
Libraries are increasingly reliant on a number of systems that are driven by an underlying ‘knowledge base’ which attempts to describe a wider world of electronic resources, enabling the library to mark those that are relevant to the local institution.
‘Link resolvers’ were probably the first systems to work this way, but latterly ERM and Discovery solutions are also highly dependent on such knowledge bases. Typically these knowledge bases do not attempt high standards of bibliographic description, but are more concerned with identifying a resource and describing how it is accessed.
Currently most knowledge bases in use in the UK HE sector are purchased as part of a suite of vendor products (Link resolver, ERM, Discovery solution) and the community around each knowledge base is defined by the customer group.
While in general the vendor knowledge bases are of reasonable quality, they are often seen as lagging behind the real world and as not doing a good job at describing the view of the e-resource world from a typical UK HE institution perspective. KB+ is a shared service for the UK HE sector, with the aim of building a community driven knowledge base, not tied to any particular vendor, but sharing information equally with all vendors and other interested parties.
KB+ has always been envisaged as a highly ‘data driven’ system, attempting to ensure that data can be created once, shared with many, and corrected easily with a single correction automatically being shared with all relevant institutions immediately.
b) High dependencies, lack of standards
Knowledge bases operate in the space between Link Resolvers, ERMs and Discovery solutions. In some cases it is impossible to use one of these products and not also adopt the linked Knowledge base. All three classes of product mentioned are absolutely dependent in their function on an underlying knowledge base.
However despite the high dependencies, there is a lack of standards for integration between these systems and the supporting knowledge bases. The most well adopted ‘standard’ in the knowlege base area (KBART) is actually a set of guidelines rather than a formalised standard, and is designed only for very basic data exchange between publishers and knowledge bases in use in libraries.
This situation of high dependencies and lack of standards can create a closed ecosystem where an institution is left with no practical alternative than to take the knowledge base and all related systems from a single vendor
KB+ offers a potential solution to this by offering an independent set of data which can be fed into any existing knowledge base without any additional dependencies, and so reduce the cost of moving between knowledge base dependent systems (link resolvers, ERMs, Discovery solutions).
c) Existing community work
KB+ is designed with community collaboration in mind. The service accepts as a premise that any centrally managed knowledge base will always be inaccurate to some degree, and that the best information about the availability of resources is held by institutions. KB+ is designed as a shared service where the community can easily contribute to the ongoing accuracy of the knowledge base, and where single corrections can be immediately shared with all institutions.
d) Financial information
KB+ has not intially attempted to manage financial information. However there is a growing demand from the initial user community for some financial information to be stored alongside data on subscriptions and licences.
3. Methods Used
a) Function Palette
b) Business Owner Map
c) Business Benefits Ranking
d) Dependencies Matrix
Scores in this matrix were allocated 0-3, where ‘3’ indicates a high degree of dependency, and ‘0’ indicates no dependency.
The scores of ‘3’ indicating high dependence of the Discovery Layer, Link Resolver and Vendor KB on KB+ are based on an assumption that KB+ is able to supply relevant data for an institution which can be used to inform other systems in use and is also feeding data to vendor KBs.
e) Shared Services Spectrum