Thanks for visiting Research Information.

You're trying to access an editorial feature that is only available to logged in, registered users of Research Information. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Web-scale aggregation promises to unlock libraries' potential

Share this on social media:

Topic tags: 

The Norwegian library consortium BIBSYS recently selected OCLC's Web-scale Management Services as its new system. Sian Harris found out why and what this means for Norway's libraries

In 1969, Frederick Gridley Kilgour, founder and first president of OCLC, suggested that ‘application of computation to effect economies will not be enough. Only by development of an innovation technology that will continuously increase the productivity of library staff will the soaring costs of libraries be brought into line.’

Kilgour served for 13 years as president of OCLC, an organisation that has been fulfilling part of this vision for many years with WorldCat. WorldCat, which started in 1971, is a global network of library content and services. It holds more than 200 million records that represent more than one billion items owned by member libraries, spanning print, electronic and digital materials.

‘There are probably 700,000 libraries around the world that run catalogue systems and most of the things that are in their catalogues are also in WorldCat,’ commented Jay Jordan, current president and CEO of OCLC. ‘It brings efficiencies through aggregation and cooperation.’

And now OCLC is also starting to deliver the second part of Kilgour’s vision: running library services on a central platform too, or, as OCLC describes it, delivering these services on a web scale.

As Jordan explained, ‘There are collections all over the world. There is a very wide array of metadata describing a broad spectrum of objects that libraries are responsible for. We need to aggregate this data at scale to get web scale.’

This vision is enabled with recent developments in computing technology, cloud computing, where computing resources from shared servers are used via the internet as required by applications. But cloud computing is only part of the new tool. It also needs to have things like reliability, security and privacy. ‘It goes beyond just running applications,’ explained Jordan. ‘Web scale looks like Google, Amazon, Ebay or Facebook. There are huge global communities of stewards, like the community of libraries,’ he continued.

Eric van Lubeek, managing director of OCLC Europe, Middle East and Africa agreed: ‘Our Web-scale Management Services is a totally new architecture. Recent developments in computing technology have made it easy to scale and WorldCat is at the heart of the system,’ he said. ‘Now with web-scale circulation management there is aggregation of data, information and services. Libraries provide many of the same services. With this they just need a browser to provide services to end users.’

OCLC’s web-scale library management tools were launched last year and are already being used and tested on a fairly small scale in seven or eight systems in the USA. There is a much bigger project on the horizon though: towards the end of last year OCLC was contracted to supply a web-scale library system based on OCLC’s Web-scale Management Services throughout Norway. This web-based suite of library management tools will handle metadata management, acquisitions, circulation, licence management and workflow for the 110 libraries, including the country’s national library, served by the government agency BIBSYS (see box).

Time to update

BIBSYS’s existing system was first developed in 1972 and has been continuously updated since then. Much has changed since then both in the resources accessed through libraries and in computing technology.

‘We thought it would be too costly and take too long to develop a replacement ourselves so we put it out to tender,’ said Roy Gundersen, director of BIBSYS. ‘We did not want to buy an existing old system because we already have an old system. We chose a system that is not ready, but will become a system for the future’

‘Usually libraries look for established traditional choices but BIBSYS especially stipulated that it wanted vendors who were working on future-looking solutions,’ added Eric van Lubeek.

The requirements set out in BIBSYS’s call for tenders were quite extensive. As Gundersen explained, ‘It is important to have easy access to electronic resources and to integrate the library system with other systems used by the institutions.’

It was therefore important that the new solution has a service-oriented architecture and that it will enable BIBSYS and its member libraries to develop their own customer applications too. These things include complying with Norway’s legal deposit requirements, the national library’s work on digitising old materials and potentially working with around 32 institutional repositories using the same system.

Roy Gundersen, director of BIBSYS

Library users also need to be able to connect to other services easily for libraries, such as their institutions’ learning management systems. ‘The most important thing is for our system to function in a way that gives students and researchers what they need,’ pointed out Gundersen.

The system that OCLC is developing should meet these requirements. According to OCLC, Web-scale Management Services ‘effectively integrate components such as acquisition, licence management and circulation with other OCLC services also operating at web-scale including cooperative cataloguing through WorldCat and discovery through WorldCat Local, to leverage efficiencies, lower cost of ownership and free libraries to spend time on unique local services and innovations.’

Scalability

Jordan is also excited about the scaling potential of the tool. ‘We did calculations on total transactions and we know that we can scale to 5000 transactions/second. There’s nothing frightening about this.’ He went on to add that having this capacity enables libraries to run analysis on the data. ‘The promise of the analytics is huge. We could, for example, see everything that was ever written by a Norwegian anywhere in the world. We could build up a body of knowledge.’

But installing such a system is not a quick process. Every library and every library system does things differently and, even when two libraries use the same system, the implementations are often very different, with different metadata and key words, for example. And the scale of the BIBSYS project is large: there are over 16.5 million copies in the BIBSYS library database and nearly 5.5 million unique titles.

‘There is quite a bit of work to do on the data,’ explained Jordan. ‘It depends on the MARC records and is an iterative process as we really want to make records perfect. Once the population of the data is complete, to start using the system will simply be a case of flipping a switch.’

This approach should make life easier for libraries, especially if, unlike BIBSYS, they all start with different systems. ‘Libraries running individual systems have to schedule maintenance and upgrades based on timescales dictated by their vendor. The threat of de-support looms large if a library is running a version deemed out-of-date. With web-scale services upgrades are all done in the cloud,’ pointed out Jordan.

The new Norwegian library system will be live on 1 January 2013. Before then the environment will be open to BIBSYS for training and testing and then for putting data in during 2012.

Gundersen is looking forward to many benefits for the BIBSYS libraries once the new system is switched on. ‘Our libraries are eager not to use so many resources on in-house work but to have more staff directly serving students and researchers instead,’ he said. ‘This will be an opportunity to rethink how the library functions.’

‘We have to streamline the process and eliminate costs so that resources can be redeployed within libraries,’ agreed Jordan. ‘If systems are inefficient we are not serving users as well as we could.’

About BIBSYS

BIBSYS provides a library and information system to Norway’s university libraries, college libraries, a number of research libraries and the National Library. It began in 1972 as a collaborative project between The Royal Norwegian Society of Sciences and Letters Library (Det Kongelige Norske Videnskabers Selskabs Bibliotek), the Norwegian Institute of Technology Library (Norges Tekniske Høgskoles Bibliotek) and the Computer Centre at the Norwegian Institute of Technology (Regnesenteret NTH). The purpose of the project was to automate internal library routines.

Now BIBSYS operates a library system for Norwegian research and special libraries and the national library. The target group has also expanded to include the customers of research and special libraries, by providing them easy access to library resources.

‘BIBSYS was founded to provide a centralised system for several libraries because it is much easier if you just have one bibliographic record for several libraries. BIBSYS libraries therefore already use a shared catalogue,’ explained director Roy Gundersen. ‘With more data you can have more interesting and useful searches for users.’