Technically Speaking

By David Dorman
American Libraries Columnist
ddorma@ltnet.ltls.org

Library consultant for the Lincoln Trail Libraries System in Champaign, Illinois.

Column for August 2001

What’s Cooking at the ALA Exhibits

Conference-goers strolling through the two exhibit-hall caverns of the Moscone Convention Center in San Francisco could sample a smorgasbord of both new and traditional information technology being offered at the 2001 ALA Annual Conference.

The surprise appetizer of the exhibit was OpenBook, a new library management system (LMS) based completely on open-source software. Developed by the Technology Resource Foundation of Seattle, the software is based on Koha (meaning “offering” in Maori), the first open-source LMS to be developed. Koha was released in January 2000 by New Zealand’s Horowhenua Library Trust and Katipo Communications Ltd.

OpenBook, which is MARC21- and Unicode-compliant, is specifically designed to achieve database synchronization among multiple servers not connected by reliable telecommunications. Willem Scholten, Technology Resource Foundation executive director and one of the primary coders, said that one of the first uses he envisions for OpenBook will be among libraries in the developing world that cannot afford commercial ILS products and don’t have the sophisticated telecommunications capabilities available to libraries in North America, Western Europe, and the developed nations of Asia.

The software runs on Linux and Windows NT, uses the MySQL database manager and an Apache browser interface, supports Z39.30 as both a client and a server, and was coded primarily in the PHP language. The development of OpenBook was funded by a variety of foundations, including the Waitt Family Foundation and Gateway Computers. Two modules, cataloging and OPAC, are currently functional, and circulation is expected to be ready this fall, after which the foundation hopes to actively market OpenBook. The targeted market will be small public and school libraries in North America as well as libraries in developing countries.

With the debut of OpenBook, open-source LMS software is poised to make a significant impact on the library automation marketplace. It may also have an impact on the politics of library automation: A fully functional open-source LMS may tempt those libraries who joined consortial automation efforts reluctantly, out of a desire to save money, to jump ship for a viable low-cost or no-cost alternative. This appetizer is mighty tasty but may wind up causing indigestion in some quarters.

Finger lickin’ feasts

Speaking of delicious appetizers, the exhibits were sprinkled with booths offering periodical title list services. Remember the days when a list of periodical title holdings was a simple thing for a library to produce? That was before the advent of electronic full-text periodical collections, whose title changes on a weekly basis made it impossible for patrons to browse all of a library’s journal titles. Well, those halcyon days may be coming back with the arrival of services that can create an online periodical title list of all print and electronic journals available to a library from all sources.

The first to market was Serials Solutions, which, I was told, had signed up over 150 public, academic, and corporate libraries in the nine months prior to the conference. The company can produce an integrated list of all the printed and electronic journals to which a library has full-text access. At this point the list is only browsable, but company founder Peter McCracken, who is a reference librarian by day and a vendor by night, said that Serials Solutions is working on title searching and on providing MARC records for each title for inclusion into a customer’s local library system. The cost of a periodical title list from Serials Solutions ranges from $900 to $3,150, depending on how many thousands of journal titles are tracked.

Journal Web Cite began marketing a similar service to academic libraries in January. “Publics don’t seem to be interested,” I was told by a booth attendant. At the moment the service only keeps track of electronic subscriptions, but the ability to also track print subscriptions is coming, along with the ability to link titles to subject headings and to add a search engine. The cost of the service, based on full-time enrollment served ranges from $900 to $6,000 annually.

TDNet was the third journal-title listing service I saw exhibiting in the Moscone Center. It seemed to be closer to a full-blown journal management system than the previous two, offering title and table of contents searching and a statistical package to monitor patron usage. Cost is between $5,000 and $25,000 annually depending on the number of unique titles tracked.

Behind all these products are knowledge databases consisting of lists of titles available from all the various online subscription services of the numerous journal aggregators such as Ovid, Ebsco, and PsycAbstracts. While Jake, an open-source precursor to these services, is the granddaddy of them all, its volunteer-maintained knowledge database of where titles are indexed is just not accurate and up-to-date enough for most libraries to rely on.

But don’t count Jake out just yet; it could still get the voluntary support it needs to maintain an accurate database. Not exhibiting, but attending meetings and wandering through the exhibit areas was Eric Hellman, president of Openly Informatics, who was handing out flyers announcing 1cate (short for one-click access to everything). Hellman describes 1cate as a “focal point for open-source development of link-server software.” Among the 1cate family of link servers is 1cate for Jake. In addition to promoting and supporting open-source software based on the emerging OpenURL standard, Hellman does open source-related consulting for vendors and libraries. Maybe Jake will get a white knight in response to Hellman’s efforts.

If these four companies represent a new food group in the exhibit’s culinary mix, Serials MasterFile is an old recipe in a new package. For decades Marlene Hurst of Farmington Hills, Michigan, has been methodically adding publishing and indexing information, as well as editorial comments, mission statements, and miscellaneous information, to 3-by-5 cards of journal titles. Last year she began converting her index cards to HTML, and the resulting database includes what is probably the most comprehensive (and delightfully idiosyncratic) publishing and indexing history of journal titles in the world, representing 27,000 titles. The cost to access this historical treasure trove is $500 for academic and public libraries and $1,000 to for-profit organizations.

Munching on main courses

Now that we have whet our information-access palates on some appetizers, it’s time to look at some of the main courses-the proprietary integrated library systems (ILS) servings. While it was no longer news at the exhibits that Sirsi is in the process of buying DRA, it was news that Sirsi had completed the sale of its UnicornECOLE division to Sagebrush Technologies. Under the terms of the agreement, Sirsi will continue to develop UnicornECOLE for Sagebrush, and Sagebrush will sell it under the Accent label. The deal gives Sagebrush a system it can sell to districts and other multi-library sites, and it gets Sirsi out of a market—school library media centers—in which the company had a very small market share. Sirsi will continue to support existing UnicornECOLE customers, but will not sell any new systems.

With so many dishes being served up, it is not surprising that the vendors are beginning to position themselves differently in the market. Sirsi sees itself getting an increasing share of revenue from content, so it views its purchase of DRA as a way to rapidly enhance a customer base to which it can market content. The company offers iBistro, its information-portal front end, to any library regardless of what ILS it uses, and it has already sold a number of iBistro licenses to non-Unicorn users.

Epixtech is taking a different approach to nourishing its coffers. The company’s director of marketing, Rob Walton, said in no uncertain terms, “We are not in the content business. We won’t be in the content business. We are in the partnership business.” Epixtech has already signed an alliance with Webfeat to market that company’s single-search interface within Epixtech’s iPAC front end, and the company is already rumored to be in discussions with Muse to provide a similar service. Muse is a product of Kate and Peter Noerr’s new company, MuseGlobal, which is already being used as a search interface for multiple resources by Innovative Interface’s Millennium system, The Library Corporation’s USeeMore portal, and My Community, a portal service to local community and government resources.

Meanwhile, Endeavor and ExLibris were having a bake-off to see whose linking service will appeal to the greatest number of academic palates. The competition even extended to obtaining development talent. Russell McDonald, one of ExLibris’s software development cooks for its SFX linking product—and as knowledgeable as anyone about using the OpenURL standard for effective linking—recently defected to Endeavor, where he is helping to bring that company’s Citation Server linking product to market. Meanwhile ExLibris, which has about a year’s head start on Endeavor in this area, is already selling SFX to libraries that use other ILS vendors for their local library systems. Endeavor hopes to do likewise with Citation Server later this year.

The d2d thing

Every ILS vendor in the exhibits, from BiblioMondo to VTLS, had a particular approach it wanted to stress to exhibit-goers. Most were emphasizing the prevailing nouvelle-information-cuisine vision of a seamless path from discovery to delivery for information seekers—the d2d solution, for those who like snappy jargon—but each vendor had its own recipe for satisfying differing customer tastes. Endeavor and ExLibris focused on linking products and digital collection management. Sirsi stressed its iBistro portal and the content it can provide through Syndetic Solutions, Cahners, and other partners. Epixtech stressed its iPAC portal and the added value its customers get from alliances it has formed to deliver services. TLC, which seems to be sincerely flattering Sirsi (by way of imitating iBistro) with its user-customizable USeeMore information portal, has decided not to offer the new browser-based version of its popular Kids Catalog to libraries using other vendors’ ILS software—which in turn prompted Geac to develop its own in-house kids catalog.

At the Innovative booth I was treated to a demonstration of catalog access through a cell phone with an LED screen, which demonstrates, I was told, the advantages of using a Web server front end that allows any Web-enabled device, regardless of platform, to access the database. VTLS stressed its partnerships and its Unicode-based language-handling abilities. In the special and corporate library market, EOSi was keeping up with the latest d2d developments, but was relatively low-key about it.

Room for dessert

There were so many tasty main dishes I hate to stop describing them, but of course we must leave room for dessert. The newest sweet treat, online real-time reference service, was being offered by a smattering of vendors scattered throughout the exhibit halls. LSSI’s Web Reference Services, the old-timer in this young field, was back at the exhibits with eShare, a product based on eGain call-center management software and customized by LSSI for online reference service support.

But the competition for providing online reference support is now increasing. Two library consortia have also customized eGain software for online reference, and one of them was marketing a product in San Francisco. The Metropolitan Cooperative Library System in Los Angeles County developed a service it calls “24/7 Reference.” Convey Systems, a first-time ALA exhibitor, was showing its OnDemand suite of call-center and interactive software products, which the company has customized for online reference and research assistance.

Drinking in the e-book scene

Although conference-goers had an increasing number of e-book-related programs to attend, exhibit-goers seemed less thirsty for actual e-book products. After generating much interest at Midwinter’s ALA exhibits, Questia did not even show up in San Francisco. But netLibrary was there, demonstrating its latest enhancements and marketing program. Ion Systems and ebrary also had booths. While the former was offering a subscription program to its entire e-book collection, ebrary was not yet offering a product to libraries and seemed to be there just to remind us that it still exists and has plans for a library service down the road.

Cooking experiments

If the exhibits can be likened to a giant information food court, OCLC can be compared to a mini–food court within it. This year its menu had a bold theme—“Extending the Cooperative”—but it had no dish to match the stature of SiteSearch, which the company announced it will not support after December 2002. However, exhibit-goers did get a chance to salivate over the beta release of the upcoming Web version of the OCLC ILL service, which is expected to be on the market in August.

One new initiative out of OCLC to watch for is the Open Names Service, which the company devoted an entire booth to, describing it as “a new venture that offers libraries linking and resolution for e-commerce and rights management.” It seems to be more of an intention to build a product than an actual product. However, its description has all the hot buzz words and seems to promise that path from discovery to delivery that we are all hoping to provide for our patrons. I expect we will hear more about this nascent Open Names Service during the coming year.

Like cooking, information organization and delivery are as much an art as a science; and even as a science, they are far from being well understood. Perhaps the most telling comment I heard in the exhibit halls was a response from JoAnne Turnbull, director of product management in North America for BiblioMondo, when I asked her to show me her firm’s preferred approach to a single interface for information discovery: “We haven’t developed a single interface because we haven’t figured out the right approach yet. How we avoid the problem that the Web has now—too many results and no control over their quality and relevance—is a problem I don’t know how to solve yet.” Building an all-purpose interface to information makes rocket science look easy!