Led by Koha and Evergreen, open source ILS solutions continued to demonstrate steady growth in 2013. These systems appeal to libraries for a variety of reasons. Unlike commercial ILS products, open source code can be accessed and altered by anyone with the expertise, enabling libraries to conduct or outsource priority development work on their own schedule, rather than wait for their requests to wend their way through a vendor’s queue.
There are also no annual maintenance or licensing fees, although many libraries that adopt an open source system will spend at least a portion of these savings on third-party maintenance, creation, and training support from vendors such as LibLime and ByWater Solutions, which primarily support Koha, and Equinox, which primarily supports Evergreen. If a Koha or Evergreen library becomes unhappy with its current support, the availability of other development houses ensures that it isn’t subject to the “vendor lock” issue faced by libraries using commercial products. It has the option of switching providers without undergoing an ILS migration. Development houses include Edoceo, Emerald Data Networks, Edusys, Lyrasis, Mill Run Technology Solutions, and Projektlink Konsult Limited.
“We’ve seen an explosive amount of interest in terms of libraries switching over to Koha,” says Nathan Curulla, executive vice president of ByWater Solutions. The development house added an average of more than five new customers per month in 2013, ending the year with 68 new contracts totaling 150 libraries. With these additions, ByWater now services 785 total installs of Koha, making the company’s customer base comparable to midsized commercial ILS vendors.
Koha was originally developed for small, single-site libraries in rural New Zealand, and the view that it is best suited to smaller libraries has persisted. Just a few years ago, the system lacked the granularity to allow individual branches within multisite institutions to manage their own circulation, set their own fine rules, establish their own patron categories, or restrict access to patron information within a single library, for example. But Koha has evolved, Curulla says, and larger libraries and consortia are starting to take notice.
“It’s grown organically to handle larger and larger institutions,” he says. “Now, it’s to the point where you can literally have a consortium of 50 libraries, and each library, for all intents and purposes, could [act] as a stand-alone system but still share the same database with everybody else.”
Patrick Jones, executive director of LibLime for parent company PTFS, Inc., which develops forks of Koha under the Liblime Koha and Liblime Academic Koha brands, agrees.
“From a functional standpoint, [Koha] stands up pretty well with commercial products that have had the benefit of lots and lots of research and development money put into them,” he says. “Three or four years ago, if I went to respond to an RFP, maybe 30 percent of the things [requested] I would have to say ‘we can’t do that, however we will build it for you if you like.’ That [percentage] has been drastically reduced in the past couple of years.”
LibLime added 30 new customers this year as well and now services 578 total installs.
Table 3: Three-Year Trends for Open Source Developers
|NEW CUSTOMERS||U.S. SALES||NON-U.S.||TOTAL|
|ByWater Solutions (Koha)||54||34||68||63||5||785|
|LibLime Academic Koha||7||5||4||0||4||132|
|Numbers represented here were reported to us by associated vendors.|
|SOURCE: LJ LIBRARY SYSTEMS LANDSCAPE STUDY 2014|
New and notable
Koha kept pace with key industry trends in 2013. Eight of ByWater’s customers agreed to collaborate and fund the OverDrive Integration Development, which makes it possible for libraries to integrate results from their OverDrive ebook collections directly into Koha search results. ByWater CEO Brendan Gallagher says that the company plans to complete similar integration efforts with Baker & Taylor’s Axis 360 ebook platform and 3M’s Cloud Library by the end of 2014. Other current projects at ByWater include an advanced cataloging module and enhancements to Koha’s federated search capabilities. In addition, both ByWater and LibLime have developed APIs (application programming interfaces) to facilitate integration with third-party products. In January, LibLime released Academic Koha 5.6, which includes a new discovery layer incorporating these APIs, along with an OPAC editor, and mobile device support.
Although Koha is now capable of supporting larger library systems and consortia, Evergreen’s scalable, client-based architecture historically made it the go-to open source option for large academic libraries and complex public systems. Recent months have seen several significant announcements from Equinox, a leading development house founded by the original designers of the Evergreen ILS.
“There were some technology-related reasons that we felt like it was a good idea to move away from the platform. But there’s also an overwhelming consensus from our users that they would prefer browser-based delivery of their tools,” says Grace Dunbar, vice president of Equinox. “It’s easier to train on, it’s more familiar.”
Subscribing to service
Separately, Equinox introduced a new suite of services at the American Library Association (ALA) Midwinter Meeting in Philadelphia in January, including two new subscription services designed to offer users enhanced support, while making the ongoing maintenance and improvement of Evergreen more sustainable. Idea Lab subscriptions will help fund ongoing development efforts led by Equinox and the open source community, and in exchange for their subscriptions, participating institutions will have input on the selection and development of features that are most important to their library. Similarly, AIM (Active Integrated Maintenance) offers three different service levels/subscription types, all of which allow libraries to prioritize bug fixes and receive immediate access to Equinox-approved fixes, much earlier than these fixes will be available through their respective community code bases.
For some libraries, these three services might help ease the transition from commercial solutions to an open source ILS.
“Coming from a proprietary [ILS] world, if you’re not an open source early adopter, there are certain assumptions built into your worldview that are hard for a peer-community effort to take care of,” explains Mike Rylander, director of development and technology for Equinox. “As a service provider focusing on these things, we’re able to dedicate resources where the volunteer community may not be able to.”
Tying these new services together is Sequoia, a new cloud-based platform that will enable users of Evergreen to dispense with local servers and have Equinox host their ILS in a fully redundant cloud environment.
“It’s a hosting and services platform that will [also] allow us to deliver AIM-derived bug fix and [quality assurance] releases to our customers immediately,” explains Rylander. “All of the code that we produce through AIM and Idea Lab are open and available to the community at large, but with Sequoia, we’ll have a platform for delivering those improvements to our customers in a more timely fashion.”
In addition, the Sequoia platform will help simplify integration efforts with ebook distributors and other vendors, he adds.
“Things that usually take some configuration and instance-specific changes to make sure everything is working right, we can do that once on the Sequoia platform, and all of our customers will benefit from it.”
As these solutions have matured and their user bases have grown, the choice to go open source has become less of a leap into the unknown. These are now well-established ILS options with many of the same features, functionality, and stability as their commercial counterparts. And some librarians feel that open source offers a greater level of control over how their systems work.
“Everyone on staff can contribute to making it better,” Henry Bankhead, town librarian, Los Gatos Library, CA, and an LJ 2014 Mover & Shaker, told LJ in a 2013 interview regarding Koha. “A clerk, a library assistant, a cataloger, a librarian says, ‘I want the catalog to work this way.’ We can make a proposed improvement, get a price for that improvement from our developers, and then ask the community if anyone else wants to contribute to that improvement…. If they don’t, we can decide whether it’s worth it to us to fund it.”
However, there are limits to this flexibility. Open source may be well suited for customization, but, at a certain point, modifications can result in a new version of the software that is no longer fully compatible with the version used by the broader open source community. One notable example of a project fork is the heavily customized version of Evergreen in operation at the King County Library System (KCLS) in Washington.
Similarly, in 2010, Liblime delivered about 70 new development features to its clients before the broader Koha community had incorporated those features into a production release. Liblime Koha and Liblime Academic Koha have continued down a separate development path ever since. Unless a library, like KCLS, is willing to pay for a high level of customization, open source development requires some degree of consensus within a community, and libraries may need to choose among different forks of similar products.
Kuali makes an entrance
Regardless, opting for noncommercial ILS solutions has become a mainstream trend, and this year will see the much-anticipated entry of the Kuali Open Library Environment (OLE) into the market. This next-generation system is organized into four separate modules, including acquisitions, cataloging and records maintenance, circulation, and systems integration, a module that will facilitate Kuali OLE’s internal integration, as well as integration with other campus enterprise systems, such as financial, identity, and learning systems.
Kuali OLE is a “community source” project, which differs from open source in that initially the software was controlled by a coalition of research libraries that agreed to contribute to the project in exchange for input regarding its development in the earliest stages. Those libraries include Indiana University, Bloomington; University of Maryland, College Park; Lehigh University, Bethlehem, PA; University of Chicago; University of Pennsylvania, Philadelphia; University of Michigan, Ann Arbor; Duke University, Durham, NC; North Carolina State University, Raleigh; Bloomsbury Colleges at the University of London; Villanova University, PA; and a consortium of university libraries in Florida. Project funding has included contributions from several of these libraries, along with grants from the Andrew W. Mellon Foundation, the most recent of which was for $882,000 to Indiana University in March.
Kuali OLE 1.0 was released in November 2013 under an open source Educational Community License. This release will allow early adopters to review installation requirements and begin mapping data from their current ILS. Version 1.5.0 was released in February and offers all of the necessary functionality for full implementation, although critical bug fixes are still being addressed. The University of Chicago and Lehigh will be the first to go live with the system in its 1.5 iteration and are aiming for summer 2014.
In December 2012, the Baltimore county Public Library—in partnership with the 3M Cloud Library and Polaris—achieved a long-awaited goal for libraries, becoming the first library to go public with a fully integrated online catalog that enables patrons to view physical book and ebook collections together and to discover, check out, and place holds on ebooks without navigating away from the catalog.
Around the same time, using the metadata, availability, and search APIs that OverDrive released in 2012, combined with screen-scraping techniques to pull data from human-readable outputs on OverDrive’s site, Colorado’s Marmot Library Network quietly accomplished something very similar with its VuFind+ open source library resource portal. That system automatically integrates each Marmot member library’s OverDrive collection into its search results, displaying print, ebook, and audiobook content together and allowing patrons to filter by EPUB or Kindle compatibility and send selected ebooks to their device without ever leaving the search interface.
Marmot’s quick rollout of an ebook integration solution shortly after OverDrive released its initial set of APIs is a good example of the flexibility and functionality that open source software can offer when backed by a good development team. And, as Marmot’s executive director Jimmy Thomas notes, opting for an open source discovery layer has allowed the consortium to enhance and customize the interface used by its patrons, while continuing to work with standard commercial ILS staff modules (currently Millennium from Innovative Interfaces). Marmot’s members seem generally pleased with this happy medium, he adds.
“When I talk to my stakeholders—librarians all across my consortium—about the open source experience, they all say, ‘We love VuFind, we love the experience. Yes, it was ragged sometimes. Bleeding edge is kind of painful. But we’re really happy with the result,’ ” Thomas explains. “When I ask, ‘What do you think about having the same experience with the ILS?,’ everyone says, ‘No, that was enough fun.’ ”
Developed and maintained by Villanova University’s Falvey Memorial Library, PA, VuFind has become the leading open source discovery solution since its launch in 2010. In addition to public library groups such as the Marmot Network and Pennsylvania’s statewide SPARK system, VuFind has been adopted by several major academic libraries, including the University of Michigan, Brown University, RI, and the London School of Economics. Core features of VuFind include a “Google-like” search experience via a single, basic search box with faceted results drawn from catalog records, digital library items, institutional repositories, institutional bibliographies, and other resources, “more like this” resource suggestions, persistent URLs enabling patrons to bookmark their queries, compatibility with Zotero research assistant tools, and more. In January, VuFind 2.2 was released with several new features, including new themes that incorporate responsive web design principles, which will enhance patron experience on smartphones, tablets, and other mobile devices.
Other open source discovery projects include the Blacklight open source OPAC developed by the University of Virginia, Stanford University, John Hopkins University, and Boston’s WGBH and SOPAC (Social OPAC) developed and maintained by John Blyberg and supported by Connecticut’s Darien Library.
Much like open source ILS solutions, working with an open source discovery interface can be rewarding, but Thomas cautions that libraries shouldn’t view these options as “free,” particularly if they are interested in customization and development.
“In our limited experience here, open source is not saving us money,” compared with using a commercial discovery layer, Thomas says. “We are getting, in our situation, way more of what we want, when we want it, for—in very simplistic terms—the same price. I used to say twice the functionality for half the price, but more carefully looking at the finances, I’d say it’s three times the functionality for the same price.”