The Future of Integrated Library Systems: An LJ Round Table

It's true, the Internet has transformed libraries. However, its impact has been even greater on integrated library systems (ILS), the technological backbone of many library operations and services. The ILS-remember when they were called OPACs?-are involved in every aspect of collection management, from acquisitions to cataloging and from circulation to web-based reserves. Going beyond the catalog, ILS vendors today strive to integrate a slew of other web-delivered services, including enriched content, shopping cart functionality, and 24/7 access.

LJ's annual Automated System Marketplace (see LJ 4/1/03, p. 52ff.) looks closely at the business climate, company standings, new products, and trends in the ILS marketplace, but the editors at LJ decided it was time to go beyond the past year and initiate a conversation about the future of ILS. How are developments in Internet technology affecting library systems? What do librarians want from their ILS? Where do industry experts see ILS going?

The Internet is opening up library collections and services, but it is also setting new expectations for ILS vendors. On the consumer side, sites such as Amazon.com and Landsend.com are also raising the bar for online services through their online help and immediate gratification. On the technology side, open source software, specifically Linux, is being adopted by major U.S. companies, providing them with both lower costs and increased flexibility-things librarians want from their ILS.

The potential to improve interoperability drives librarians to look more critically at ILS. Some wonder whether the information portal of the future should be based on a single ILS or instead be a collection of products from different ILS vendors. This piecemeal approach to interoperability in the library marketplace has been created by the vendors themselves with such products as Ex Libris's SFX, a tool for reference linking, and Endeavor's ENCompass, a product for creating and managing digital content. Standalone products for linking and digital management accounted for nearly 13 percent of the ILS market last year.

Metasearching, the ability to search multiple resources at once, is a powerful new capability that gives librarians the freedom to determine how they want to provide access to their catalog and database content. Libraries acquire metasearch engines through their integrated library system vendor-most provide it through partnerships-or direct from companies that specialize in metasearching.

Internet technology is also fueling librarian initiatives nationwide. From moving the catalog to the surface of the web to finding new solutions to interlibrary loan, from creating digitized collections to managing e-reserves, librarians are creating solutions that often stand outside of the ILS. Will the ILS be the glue that holds together the digital library of the future? What follows is a distillation of two hours of discussion.

Fialkoff: What industries or companies do you look to for clues to innovation or directions you need to be thinking about?

Grant: I'm sure we all follow Microsoft, whether you love them or not, and Amazon.com. We watch companies like Oracle, too.

Walker: Dropping down a level, at the moment I am focused on the semantic web, some of the Internet2, and the Middleware projects. And also on some of the federated network ideas on the web that provide a single sign-on, so you can actually sign on to one system, such as an airline system, and then go directly to a car rental system.

Blauer: We also look at a lot of different companies, across telecommunications, what's happening with personalization.wireless applications. From a big-picture standpoint we're watching the whole area of technology convergence. Those things happen in the general world a lot faster than in the library world.

Goldner: We follow emerging technologies because it's what's under the hood that's always been of great interest to our company. Sometimes that works out really well. You jump on something, you try it out, and that goes. Occasionally, you jump on something, and you are 12 months down the road and no one else has gone in that direction.

Kline: What has been left out here is the whole service industry. Because our customers are service customers, we also follow how companies provide excellent service to their end users, giving the customers what they want. [The service industry] is a very important industry for us to follow, whether it is the banks or the retail industry or whatever. The same people who use the library use banks and buy clothes.

Wilburn: People come into the library and bring their perceptions of different commercial sites, like L.L. Bean. They want the library to look like that. They want to retrieve all sorts of information and don't care how it gets there.

Sommers: What is competing for people's screen time? That's where the library has to look to understand what people like about interfaces, what people like about other types of systems. More and more, access to library resources can be done through an interface. The reality is that the future's here, it's just not widely distributed yet in the library industry. No one brought up the entertainment industry, but I think a lot of the technologies in gaming, the way people are building new interfaces, the way people are getting at information are very interesting. We need to focus on making the experience better for the end user.

Blount: Two of the companies we track the most are Sun and IBM. IBM is really leading in the area of Linux, which is going to be a very important part in the future. Certainly Sun and Java also. Everything we do in development today is done around Java and J2EE, standards that didn't exist ten years ago. The third area that wasn't mentioned is the W3C, which is an Internet standards organization that we're a member of and participate heavily in because we think that's where most of the new standards [will emerge].for interoperable yet secure and robust [systems] that will allow integration transparently.

Kenney: Google and Amazon keep coming up as models for online services. How are you responding?

Goldner: Our first step was to solve the Google issue. The difference, as a library vendor, is that [the results] have been categorized and contextualized by the librarians ahead of time. So that when people authenticate, we know what type of individual they are and therefore we have preselected. We've set up targets for them, if they just come in, want [to do] a fast quick search and get out.

One of the things that the ARL Scholars Portal Working Group wants is a panic button on the screen for students who've walked into the library at night with a paper due the next morning. Because the students will have authenticated, or can if they choose to, we'll know they are English majors. The panic button will actually search certain databases automatically. The interesting dilemma we face, and as a librarian I think I'm allowed to say this, are the librarians. They often want more complexity in the interface than end users want.

Blount: We really have to focus on simplicity. We're all recognizing that that's what users want, but you can't stress enough the authoritative content. On the Internet you can get completely conflicting information, whereas in the library environment when you're searching.you can still find any information you want, but it is authoritative.

Ludwig: We've constructed our own web-based catalog using XML technology and a full-text search engine from a vendor on top of that. We're not really providing a library system, we're providing library content on the net. We realize we have to compete with Google, and students and faculty are now [more familiar with] that interface than any library interface or database.. If you look at the roots of many of the systems represented around this table, they're all evolved from very old systems. We have to deal with text as pages that people can retrieve in a much larger world, much larger web environment.

If you look at our [academic] libraries, we're concerned with electronic resources. We buy very few books anymore; our serial subscriptions are up. We're buying electronic content; that's where our money's going. The library catalog has to fit into that model. With Amazon, you get pictures, you get reviews-many of us have linked library catalogs to Amazon. But we can pull that same content into our own catalogs using the XML standard content. Syndetic Solutions provides a database of this content. Once you begin to get out of the old system construction and use the modern web technology, you get the advantage of integrating more and more standard web objects into your system.

Blauer: Some of us at this table have taken that leap into new technologies. We're not based on those old systems-although we might have a legacy system that we're migrating our customers from. We're using XML, we can catalog the web on the fly. We're using XSL style sheets. These are not just add-ons, these are integrated right into our product.

We all work with limited funding, including libraries and the vendors and all of those other companies that we're trying to integrate. So it's better if we work together to develop some of those things, then institutionalize them and have products that are scalable. By working together we can do more than if we're working in isolation-in essence, competing.

Some companies like us have faced the challenge of having a totally new platform. What comes back to haunt us is that we're building the functionality that librarians expect of third-generation automation systems, but they don't want to buy it because it doesn't have every bell and whistle for all the backroom functions that they're used to.. We're all trying to do critical things with the technology to serve better the 170 million folks on the web right now in the United States. What percentage of those people are coming to libraries?

Grant: What I disagree with Mark [Ludwig] about is that it isn't just a matter of pushing our content to the surface of the web, it's pushing the value-added services of librarianship. Content is out there; the problem is how you filter through it to get to the important information.

Kenney: At a presentation yesterday, a librarian was discussing ILS, and she proposed that we should stop talking about integrated systems and begin talking about integrating systems-that is, libraries buying different modules and integrating them themselves.

Dietz: I've been in the computer industry for a long time, and I can tell you it is nice in theory, but [buying] a lot of programs and [integrating] them make systems more and more difficult to support. It can't be done just by the librarians or by the library organizations. They need some kind of support, and at the end of the day a lot of our customers will come to us.

Grant: I totally disagree. The functional integration of systems is absolutely where we are headed. People are going to want to hook up different pieces of software and expect them to work.. There are libraries out there that are using PeopleSoft as an acquisitions system. There's going to be more of this, and we have to get ready for it.

Blount: Integration is the wave of the future. If you look back at the 1980s and the 1970s, we had proprietary computer systems. The databases were proprietary, the operating systems were proprietary, the hardware was proprietary. You couldn't integrate anything. I think it is proven that open standards and integration bring you lower costs, higher performance, and more robust solutions. You go into anybody's system today, [and you'll see] they're not running only IBM software or only HP software. They're running an integrated solution of multiple vendors through standards like TCPI-which everyone said wouldn't exist or come about. It's the reason we could have an Internet network today.

Walker: Librarians want to choose the best product. I think it's healthy to get away from legacy systems that put these products like a federated search tool over a system.. We've talked about interoperable systems for many, many years now, and we're finally hearing librarians say this is what they want. I think the key thing for them is a single point for data entry.

Goldner: Librarians don't want all their eggs in one basket any more. They want to be able to pull things together; that's what metasearching has brought up. It's the first place where librarians can actually buy from several vendors and have the products work together extremely well. Years ago, I heard librarians saying, 'Someday we'll be able to buy our acquisitions module from that vendor, our circulation module from this vendor, our serials module from that one.' It just didn't happen. How libraries buy and implement systems is really changing.

Blount: The challenges of system integration are that there just haven't been enough resources out there. Who pays for that system integration? If you're a big corporation, and you buy a CRM [customer relationship management] system, you probably realize that what's really going to cost you is the system's integration, not the CRM system itself. Where we face a challenge is that we're trying to integrate many things, because I still see the ILS as the core of what a library does.

Fialkoff: We've heard a lot of complaints from vendors and librarians, about how arduous and costly the RFP [request for proposal] process is. Is it time to jettison the RFP process and replace it with something simpler?

Goldner: I wish we had a NISO committee for RFPs, because if there is a list of 333 standard pieces of functionality that every library wants, why can't there be a standard list-here is the beginning of the RFP-and then we just go through and check yes or no. Then they can have the five or ten pages of things that are specific to their community. Since I've joined the vendor world, I've thought it's absurd the horrendous sums of money that all of us are spending to answer all of these RFPs.

Ludwig: Some of you may be surprised to hear that I sympathize with you.. The art of RFPs is horrendous: it's horrendous to write them, it's horrendous to read them, it's horrendous to respond to them. But it's required by state law. It is also required when you're dealing with big consortia.

Kline: All the requirements in these different RFPs just push us to make better and better decisions. Libraries are different; I think we should be celebrating the differences among these libraries. Different communities, different colleges, feel different about what their library needs to be, and they're going to put out different requirements. It keeps our industry vital.

Blount: We are seeing a dramatic change in RFPs, which we actually find exciting. The last three RFPs that I've been involved in have had less than three pages of traditional backroom functionalities. They wrote 14 pages on digitization and 22 pages on the information portal. That's part of the uniqueness of this market that makes it system software, not off-the-shelf package software. Every library is a unique environment.

Kenney: In LJ's Automated System Marketplace this year, we predicted that migrations from legacy systems over the next three years will really shape this business. What do libraries risk by delaying migration?

Dietz: The budget situation is not very nice, but at the same time, every time I speak with a library director one point is stressed: they are looking for efficiency in the back office functions so that people can be routed to be more with their users. Several directors told me they would expect us to be change agents.

Blount: What do you miss, what do you give up by not migrating, by not staying current? That's a whole catch-22 that libraries sometimes forget. They exist and grow based on public funding. They get that public funding based on the services that they provide the community, which means Internet browsing from home at night, authoritative data from hundreds of databases, digital access to rare book collections. If they're not stepping forward with new technologies, letting me use my PDA when I walk into the library to do my search, then they're going to lose their patrons and their funding. The whole cycle will collapse.

Wilburn: I totally agree. One thing that has caused increased usage is the look of our OPAC. The end user comes in and sees our new catalog, the enriched content, and that creates excitement in the community. People come in and want to see what we have. We get improved circulation from it, and that turns into money.

Walker: Libraries that are stuck with legacy systems do have options. People can now purchase a library portal or information gateway, layer that over a legacy system, and immediately expose information to patrons and enable integration with electronic resources and digital content.

Walker: The main impact of meta-search engines is on the OPAC. It will be very important for libraries since it can expose so many library resources.. There is no one single route, so even a metasearch engine is just another route to the information to make it easier for users. There may well be one route through the Internet, another through the metasearch engine, and another directly through the native interface. We as service providers have to allow for all that.

Blount: I think it's clear that the people do want metasearching integrated into their PAC [public access catalog]. It comes back to the issue we talked about earlier: users still want it to be simple. They get to a lot more and a lot richer content, but they want it to be simple, and that doesn't mean jumping around to 20 different interfaces with different search engines.

Kline: The ILS is not a stagnant thing. It is continually growing, with new pieces always being added to the puzzle. And federated searching is just another piece of that puzzle for providing services for libraries.

Blauer: Many of us spend an awful lot of time looking at how to improve the end user experience. That's why we look to metasearching portals. How are we going to get the folks who visit Terri's [Wilburn] library the information they need? The mantra of public libraries right now is just for me, right now. I want content served up to me. You need to know who I am. I don't want to be logging in. I don't have to be fishing in my pocket for a library card. I don't need to do that on Amazon. You need to know me as soon as I get into your system, the first interaction I have. And push it out to me as well.

Kline: We have to look at how metasearching is affecting ILS. Like interlibrary loan. Once you have a federated search engine finding information, you will need to move more information between systems. Digital Rights Management, in case it is e-content you need to move around.

Fialkoff: What are you working on right now that you think is the most innovative? What are you most interested in at this point?

Blount: Looking to the future, one of the most important questions for our industry is the role of Linux and open source software. Linux will be one of the most important operating systems in our market in the future, and a lot of that is driven by the flexibility and the cost. We just won a deal recently where a traditional, proprietary UNIX-based server and software system was replaced with a Linux-, Intel-based server-a $60,000 system is replaced with a $6000 system-with better performance, cost savings, and more flexibility. It's a way to get back to the cost issue, to still have funding for new technologies [like] digitalization and federated searches, areas where you get more benefits for users.

As technology vendors, we have two focuses: the end users and the staff. That's our job, to bring very complex technologies together behind the scenes so that users and staff get benefits.

Grant: The real focus for us as vendors is taking the values of librarianship and pushing them out to the desktop where people are using the library. That's the most important thing we do; everything else we're doing is largely background.

Dietz: The key focus is [integration] with such functions as courseware and the visibility of the system. Today we have rudimentary federated searching. How usable is it? How can we make our systems proactive? How can we anticipate user needs? That is key.

Klein: We didn't touch on the management of electronic resources, which is a huge headache for every library in this country. We're trying to grapple with that, as well as better resource sharing for the wonderful resources that are not available electronically, those wonderful resources already in libraries: books, the paper journals, and moving them around. If you make it easy for students, they will use these resources. If it's just standard ILL, they'll never use it.

Wilburn: We look to our vendors to bring us the latest things, but we also want to be involved in what they are. We don't want to be presented with something we don't need or want. What I see coming is that people want everything really simple, just like going to a bank ATM or buying something from a catalog. I can't emphasize enough the support we need from vendors.

Sommers: What's the role of the library going to be? Impacting that is the definition of what information is. In a lot of respects, libraries have abdicated their responsibilities to the Internet.. How can you improve the experience of the user? It's a battle not for the people our age but for those people who are five and six years old. They're going to grow up with a whole different perspective. The future of libraries 20 years from now is being set today.

Walker: The focus has to be on the ease of accessibility to a whole range of library resources and the increasing range of resources. Another key factor is improving the workflow for libraries.. As you increase the resources, the management of that-dealing with intellectual property and digital rights management-will become a big issue.

Blauer: How can we enhance the end user's experience? If we're not spending the bulk of our time worrying about that, then we're not paying attention. We are deeply involved in integrating technologies with the ILS in cooperation with some of our customers because they don't have the wherewithal.

Bakhtiar: We work with our libraries and partners to develop new initiatives. One that we are working on is the acquisitions that we introduced at the winter [American Library Association conference]. We are also working on systems that will take advantage of the open source, giving the libraries their choice of where they want to go. Then, of course, personalization, YouSeeMore, so the end user can set up a profile.

Ludwig: We believe we've found a revolutionary way to integrate all of our content: our full text, our library catalog pages, our image collections, our sound files, our course reserves. We are integrating them using native XML, with a search engine to bring it all together into one common search interface. We're looking at a whole new paradigm for treating our collections as a collection and not as separate entities. We are working with IXIASOFT, a Montreal company, with a product called TextML that allows us to create XML content from almost any digital object. We're very excited because the cost is low and the functionalities are high. We're expecting great things from this direction.