You are here


Dewey or Don't We

Today's library automation systems promise to improve efficiency and save money. Turning those hopes

Michigan's Waterford School District is at a crossroads. In 1995, administrators purchased an Integrated Library System designed to automate and link Waterford's 21 school libraries and its single collection of videos. After a year of implementation and training, the district bid goodbye to its paper card catalog. In its place: a shared electronic system with a single catalog and single borrower database.

Although the ILS has been a great success, "it's behind the times," says Nancy Smith, library director for both the district and Waterford Township's public libraries. "The current system is textual with no Windows or Web capability. The students and staff are used to a Windows environment."

Ironically, the Waterford district is already ahead of the curve when it comes to library automation, simply because it uses a centralized system.

You don't throw out the librarian with the card catalog.

But it's also representative of most school districts in that it relies on outdated software. So says Gail Lewis, senior database services manager for Sagebrush Corp., which offers Accent, Athena and Winnebago Spectrum for K-12 libraries. "Our latest research shows that about 94 percent [of school libraries] are automated," Lewis reports. "But most libraries are automated on older software, or they're using software in a distributed or site-based system. Ideally they want a centralized system."

"In the past, it was traditional to have one server at a building and a library automation package that worked just at that building," echoes Carol Simpson, an assistant professor of library and information sciences at the University of North Texas. "The emerging trend is that a lot of districts are going to an all-district platform with one database for all the schools."

Welcome to the 21st century school library system, where collection, circulation and patron information are shared between buildings; where money is saved on duplicate purchases and IT staffing; where students, teachers and media specialists alike can work faster, simpler and more efficiently.

Central Intelligence

Most library automation software falls into one of three main categories:

DISTRIBUTED SITE-BASED SYSTEMS. Each library operates independently of others in the district. These systems tend to be the most cost-effective, as they don't require a district-wide network, and they can be deployed on a piecemeal, as-needed (or as-affordable) basis. On the other hand, these systems require a computer server at each library, meaning more hardware and software for maintenance and the need for more IT staffers to maintain them. Plus, software upgrades must be done on each machine, one at a time.

CENTRALIZED SYSTEMS. These systems raise the stakes with far greater benefits--and higher costs. Key among the former is shared data. Each library's collection, circulation and patron information is stored in a single database that's accessible from each library. Individual classroom access and even home access are available from vendors, although these features aren't typically part of the base price.

When a new title comes in, it gets put in the system once and only once. If it's checked out, any librarian can determine who has it, when it's due back and where. "Resource sharing at the level provided by a shared library system is not possible any other way," Smith says.

Lewis points to the advantages of consistency for students using a centralized system. "As they progress through the grades, they're using the same system," she says. "Library staffs don't have to teach students over and over again."

Although a centralized system requires just one server, the district must have a wide-area network, which can be costly to deploy if not already in place. There are also potential costs associated with training, collection conversion, system implementation and various other startup necessities.

The initial costs may be high, but total cost of ownership will ultimately be lower, Lewis says. "Centralized systems save money over time. You don't have to send an IT person to each school to install software, for instance, because all the updates are done on a single server."

"Under our old system, we actually had to create a stipended position just to handle library system maintenance," says Al Chickerneo, systems administrator for Deerfield (Ill.) Public Schools District 109. The Chicago-area six-school district recently replaced its text-based library automation system with Destiny from Follett Software and discovered that upkeep is much easier with a single point of maintenance.

A centralized system, which tracks not only book collections but also all forms of multimedia, can eliminate unnecessary duplicated purchases, too. "More than once, I would pay $300 for a math-skills [software] kit," recalls Simpson, a former facilitator for library technologies at Mesquite (Texas) Independent School District. "A while after that, I would discover that another building had the same kit. I just spent $300 on a kit the district already owned."

Yet another benefit is the reporting capabilities of these systems. "A centralized solution ... simplifies and strengthens reporting. It's easier to bring data together to create useful reports," says Kathy Sharo, director of marketing for Follett. For instance, if an administrator wants to understand why some school libraries are used more than others, comparing the usage of collections across the district is a simple process.

"UNION" SYSTEMS. Somewhere in between distributed and centralized systems lies the union system, which can be regarded as a kind of stopgap solution for districts that either can't afford or don't need a full-blown centralized package. These systems are essentially upgrades to distributed systems that enable collection sharing between libraries.

A database contains collection information for all the libraries in the district, meaning students and teachers can search the entire district from any school or Web-enabled PC.

Collection information isn't updated in real time, but rather downloaded to each school's server at scheduled times. While you avoid the high costs of a centralized system, you still have to pay to maintain and update the distributed system.

Regardless of the type of ILS you choose, the big trend in library automation is the meta search. "You search one time and you're searching multiple resources with it rather than just within the library," Lewis says. Those resources can include not only other libraries within the school district, but also the city's public libraries, the Internet and online databases (such as EBSCO and SIRS) to which the schools subscribe. "They want their staff and students to be able to search the libraries anytime, anywhere," says David Hale, a senior support analyst at Chancery Software. Time-saving capabilities like these are a chief reason Nancy Smith is eager to update Waterford's system.

Decisions Decisions

Measure twice, cut once. That's the rule carpenters follow to make sure all the two-by-fours end up the right length. It's also a fitting analogy for choosing an ILS: measure your needs carefully, making sure to consider both library and IT requirements, then cut a check. (Okay, you may have to cut more than one, but you get the idea.)

Indeed, if the motto in real estate is "location, location, location," in library automation it's "research, research, research." That's the consensus of product vendors and library personnel alike. "It's a complicated project, so good planning is critical," says Carol Paul, district library media coordinator for Douglas County (Colo.) School District.

"You need to understand what it is you want the system to do," says Simpson, who edited the recent book Ethics in School Librarianship: A Reader (Linworth Publishing, 2003). "If you've never been automated before, you may not even know what [these systems] can do. Her suggestion for the next big educational technology conference: "Wander around. See what the different vendors have to offer." Even Lewis from Sagebrush advises administrators to play the vendor field.

Paul suggests talking with individuals in other districts who have experienced the ILS purchase process. And when evaluating various products, take the time to see demos and try each one.

She also suggests enlisting assistance from the software provider and creating a long-range plan that includes a budget, as well as negotiating for discounts. "Sometimes you can connect with what another district or your public library is doing and receive a discount," she says.

Maggie Bowden, executive director of instruction for Monroe County (Ga.) School District, says the ILS purchase decision there took nearly a year. "Take the time to work through what you want first," she recommends. "Then compare [ILS] systems to that outline. Everything looks good until you have something concrete to rate it by. You can't be locked into 'That's the way we've always done it.' You have to be able to look at things in a new light."

Bowden says Monroe County settled on Sagebrush's Accent system "because we thought it was the most versatile. ... We know enough to know that we don't know what is around the corner in terms of technology.

We wanted to go with a product that would allow us to add or change policies and products as the need arose."

While you're shopping for an ILS, "beware the promises of over-enthusiastic salespeople," warns Simpson. "One district was given a terrific deal on a new software product that ran on a very specific main computer. Once the software was purchased and installed, they discovered that it had never been used in a school before, and there was no module to conduct the required inventory. The search interface was designed for adults, not elementary children. While those features were eventually delivered, it took several years to get them installed and debugged. Software specifically designed for schools would have come with those features installed."

Another potential pitfall that Hale advises districts to avoid is purchasing software that doesn't allow for portability if a different automation program is implemented in the future. Chancery's Library Pro is able to export standard MARC format files, so customers aren't locked into a proprietary data format, he adds.

Specialist Ed

There's more to choosing an ILS than just determining the most suitable system for your district or negotiating the best deal. There's also the matter of educating, training and reassuring media specialists, who may consider a high-tech system unnecessary or feel threatened by it. The experts agree: To keep them happy, keep them involved. And keep them around.

"Allowing all the media specialists to work together and talk the whole project through was a great strength," says Bowden. "They have a great deal of buy-in to the product and what we are doing. I didn't have to 'sell' them anything, and they understand the reasons that certain policies are in place."

Hale agrees. "I think one of the most important pieces of the purchasing process is imparting a sense of ownership to those who will be using the software," he says, adding that senior library staff will have valuable insights on what the software needs to accomplish for them.

"Many administrators feel that once a library is automated, there is no more need for a librarian," notes Simpson. "A library clerk can wand books in and out quickly enough. That is a short-sighted approach. The librarian is trained in the intricacies of catalog practice, federal regulations on interlibrary loan, policy development and technology management."

"No matter how easily a program allows you to enter information, the most important function of any database is retrieval. Simple searches are intuitive, but more complex searching requires some training and experience to perform successfully," says Hale.

In other words, you don't throw out the librarian with the card catalog.

That said, "Librarians are sometimes very resistant" to automated systems, Simpson says. "Before, the individual librarians were responsible for their own databases, their own inventory. But you have to learn to give up a little control."

Tales from the Front Lines

So, once the automation system has been chosen, the funding secured and the librarians reassured, what comes next? Usually a fair investment of time to obtain and implement the system--a year, 18 months or perhaps even longer.

Don't forget to implement a data backup system, as well. This is an "important process that is often not recognized as such until too late," Hale says. Data recovery can cost up to a dollar per record, he adds.

That's money, of course, that most districts don't have. "It took about three years to fully automate all our schools," says Paul. "We did not have a large sum of money to support a quick implementation, so we brought schools up as money and time allowed."

Unfortunately, not all was smooth sailing for Douglas County. "We had several instances of complete orders of books being misbarcoded with new libraries," Paul admits. But there was a silver lining behind the barcode cloud. "Parent volunteers helped with the rebarcoding. These individuals ended up being the biggest supporters of our libraries, spearheading fundraisers, which added many volumes to our collections," she says.

In the end, even the most frustrating startup hassles are usually worth the effort. Nancy Smith, who's stuck with her district's nine-year-old system for now because the funding isn't there for a new one, wouldn't run a library without centralized automation.

"The benefits to the [school] community to have a catalog at home are obvious," Smith says. "Less obvious is the fact that library automation allows a library to run more efficiently, providing more bang for the taxpayer buck."

Rick Broida is a freelance writer based in Commerce Township, Mich.

Related Information