With U of Tennessee Rollout, OCLC's Cloud-Based ILS Enters Early-Adoption Phase
By David Rapp Jul 21, 2010Little more than a year ago, OCLC unveiled its ambitious cloud-computing-based integrated library system (ILS) project, Web-scale Management Services (WMS). That project has now moved into an early-implementation stage at a handful of libraries, including the University of Tennessee at Chattanooga's (UTC) Lupton Library, marking the last major phase in WMS's development before a planned U.S. rollout next year.
WMS aims to use several existing OCLC services--such as WorldCat Local, which is the discovery layer of WMS--as well as newly developed applications, to combine the many different functions of a locally installed ILS into one cooperative online network.
When WMS was first announced last year, it generated mixed reactions from vendors and librarians, some of whom were wary of OCLC further expanding its influence in the library market by moving into the ILS sphere.
UTC rollout
After a preliminary pilot phase involving a small group of academic and public libraries, UTC and a handful of other as-yet-unnamed libraries are currently at work on an early, full implementation of WMS. On July 15, UTC heads of library technology Jason Griffey and access services Colleen Harris wrote on their respective blogs about the plan to go live with WMS by August 30. (There's also a Lupton Library wiki laying out some of the details of their implementation project.)
ILS "epiphany"
LJ talked to Executive Director of OCLC Networked Library Services Andrew Pace, who is heading up the WMS project. In 2004, as head of systems at North Carolina State University, he wrote an article for LJ about what he saw as the need for better interoperability between the various modules of ILS systems. He's been consistent with his message ever since, with a singular focus on reshaping the ILS.
Back then, "the problem was: we could take apart the ILS," he said, "we just didn't know how to put it back together again, and make it different than it was." Since then, Pace said, he's seen efforts to do something different with ILSs either fail, or merely rebuild while still retaining old legacy workflows and functionality. At OCLC, however, which he joined in early 2008, he had what he called "an epiphany": that an ILS could live on the network, in the cloud, with the power of the cooperative behind it. The thinking is that such a model could make for a lot more efficiency, and could potentially allow for easier scaling as a library grows and changes over time.
From pilot programs to early adoption
Three main WMS modules were developed: license and subscription management, circulation development, and acquisition and workflow. The circulation prototype debuted in July 2009, and several libraries--including Pepperdine University Libraries in Malibu, CA, the Idaho Commission for Libraries, and the CPC Regional Libraries group in North Carolina--tested it as part of a pilot program to make sure it provided the functionality required. OCLC user-experience experts also visited the libraries to analyze workflows.
In this latest, early-adoption stage of the process, UTC, with its ambitious and willing staff leadership, is a good test case, Pace said. "They're the exact kind of library you want crowdsourcing this kind of problem." They have been "sharing their information in a very transparent way," Pace noted, "so that we can scale this operation for [other] libraries."
What's more, the ambitious timeline at UTC is a key element. "The goal is that at the end of the adoption period, we would have our piece of this [implementation process] down to 30 days," Pace said. Other aspects, such data extraction and preparation would be handled by the libraries on their own schedule, but the aim is that the whole process would take about six weeks.
Costs and interoperability
As far as costs, the traditional model for an ILS is that the library pays a large up-front licensing fee for a piece of software, and then pays a maintenance fee for that software every year. WMS would have no licensing fee, and would instead use a subscription model, akin to other OCLC services. There would also be an implementation fee, Pace said, which would be a percentage of the annual subscription fee.
WMS comprises three different modules focusing on different ILS aspects. But can the modules work separately, or are they a package deal? For example, if a library just wants to use the circulation aspect of WMS, could they integrate that into their existing ILS? Pace told LJ that while it's technically possible, "it's not the plan for OCLC to support that." He added, "Using one system to acquire something, to get something in the door, and using another system to get it out of the door--this is not a cost savings for libraries."
That said, Pace pointed out that WMS aims to work in tandem with third-party processes when needed--some libraries, for example, are locked into using specific business processes. "Let's interact with other business processes as much as we can," he said. "But let's not build a Frankenstein's monster of inventory control."
New developments
WMS also intends to make it easier for staff developers to create new applications for the system. UTC's Jason Griffey blogged about a plug-in he'd seen that uses live New York Times bestseller list information in conjunction with the acquisition module, to allow single-click ordering.
Pace said that he believes that a cloud-based ILS model will eventually become the standard. "Do I think the other ones will fade into history in the next five years? No," he told LJ. "This is primarily about providing a new choice."
With this new phase, more information is being made public about WMS particulars. Pace gave a presentation about WMS at the Public Library Association conference in March, and he will demonstrate some aspects of WMS in a live web presentation today at 2 p.m. EST.
