OCLC's internal code name for the project is Extended World Catalog, which builds on the mammoth bibliographic WorldCat database—covering books, serials, monographs, videos, films, etc.—that it has constructed out of the holdings of its member libraries. The OCLC executives I interviewed assured me that the new strategy has received strong support both inside their governance committees—as indicated by the recent additional funding voted on by the board—and from canvassing directors of member libraries. According to Frank Hermes, vice president of marketing and planning, OCLC has established a broad team of people over the last 6 to 7 months. The program has a 3-year projected timeline, with its first efforts expected by January 2001 and a beta launch next summer.

OCLC's new strategy applies some of the concepts from the Coffman articles ("Building Earth's Largest Library: Driving into the Future," Searcher, March 1999, http://www.infotoday.com/searcher/mar99/coffman.htm; and "The Response to ‘Building Earth's Largest Library,'" Searcher, July-August 1999, http://www.infotoday.com/searcher/jul99/coffman.htm). For example, the new design plans for the Extended World Catalog include both enriched records (adding cover art, reviews, tables of contents, etc.) and expanded outreach into merging catalog data from other sources. According to Hermes, OCLC also plans to make the database "more robust, broader, and deeper, and linked to outside databases. No longer will all data flow from the computers in Dublin, Ohio. We're moving beyond bibliographic records, adding tables of contents, cover art from book jackets, more of Steve's ‘big, fat records,' through full links to digital objects, special collections, and alternative formats."

Hermes also indicated that OCLC would link Extended World Catalog data globally—e.g., with the PICA online cataloging service in the Netherlands, in which OCLC recently acquired a majority interest. PICA helps libraries in the Netherlands, France, and Germany create union catalogs. Under the new strategy, OCLC plans to link data from PICA into the Extended World Catalog and vice versa. This policy would expand both the number of bibliographic records and library holdings. No matter how diverse the sources, however, the project managers at OCLC intend to present patrons with a smooth, integrated presentation of data.

The Extended World Catalog forms just one part of OCLC's new "Four Corners" strategy, according to Gary Houk, vice president of metadata and content-management services. The basic four elements of the strategy are the following:

• Metadata—Formerly called cataloging, but now expanded beyond the traditional OCLC records to new sources from a variety of partners and even some pre-publication metadata, all designed to serve the end-user and the librarian

• Content Management—Will enable OCLC to help librarians manage their local collections, including archiving and digitizing local collections

• Discovery/Navigation with the next generation of reference services, such as the Portal Management Service—Will help librarians create their own Web sites and portals, as well as effective interfaces for patrons dealing with the Extended World Catalog

• Fulfillment—Rapid information-delivery services, including an integrated "Click to Borrow or Buy" feature. When users find a citation, they can get the item. The system will offer three layers of fulfillment options: the local library, a specific list of libraries chosen by the local library (such as a consortium's members), and/or a general custom list.

The new strategy would also help OCLC expand its presence with other Web services. Houk noted that OCLC was constantly being approached by firms looking to tap into its data store. Often, such firms have something to offer that OCLC's customers could use, such as Preview Port with its collection of author information. Houk foresees a button at Amazon.com's site that might read "Go to the Library." Houk also foresees a day when librarians can get revenue back to their libraries from these services (through referral fees paid by online booksellers).

Clearly, there are lots of problems to be worked out. For one thing, fulfillment plans must depend on access to circulation or shelflist inventory data. At present, OCLC doesn't support circulation service, leaving that to a network of OPAC (online public access catalog) vendors. When I asked Houk and Hermes about this missing key element, they recognized the problem as one of the issues that they would have to face. At present, they hope that a new ISO protocol to standardize data from circulation control/shelflist services will allow their computers to query other computers quickly and easily. Payments could also come from other third parties, such as advertisers or sponsors, as well as from patrons.

Ultimately, according to Houk: "We intend to make libraries a major presence on the Web, using the Extended World Catalog. We want to make library holdings a daily part of the Web experience for users."

Exciting visions, but what about that 3-year timeline? As I pointed out to Hermes and Houk, 3 years is an eternity in the Web world. Hermes admitted that at this stage the company is still validating and setting the framework for the vision. OCLC has a multiplicity of governance units and a process of acquiring consensus from talks with libraries. On the other hand, both executives indicated that the commitment from the organization has gone way past the "nice idea" stage.  They hope to have some of the initial work available for view before the end of this year.

The problem, as Hermes put it, is that "we are a dot-org institution trying to work at dot-com speed."

Initially, OCLC will set up four library consortia for testing the new Extended World Catalog and Portal Management Service.

I asked Steve Coffman, now employed by LSSI, the country's largest library outsourcing agency, whether the ideas in Extended World Catalog looked like what he had in mind when he wrote the Earth's Largest Library articles. He responded, "[It] looks a lot like it … I wonder where they got the idea?" He indicated that the beta test would probably involve a select number of key library consortia and would result in the release of something impressive out by early next year. Overall, Coffman noted that the new program was "moving remarkably fast for OCLC."

As for the problem ahead, Coffman also pointed to the circulation records as the "missing piece, essential for delivery times and the placement of single orders" with the right library. OCLC might face particular trouble, according to Coffman, if local OPAC circulation system vendors don't want to buy into the ISO standard data-sharing. However, he suggested that OCLC might want to look into partnering with low-cost, Web-based circulation system providers using the new Application Service Provider (ASP) model. In all the reactions Coffman received to his article, good and bad, the one universal sentiment was dissatisfaction among librarians with current OPAC vendor services.

In general, Coffman stated: "OCLC likes to see itself as ‘one of us' [librarians], but many librarians don't see them that way. Some remember being left high and dry when OCLC pulled out of earlier projects. Some think of OCLC as just another vendor, only slower than most. But OCLC should prevail because it would be almost impossible to replace that incredible asset of the World Catalog."

OCLC's basic WorldCat database contains over 43 million records and grows at the rate of over 1 million records every 5 months. The initial Extended World Catalog will also offer over 10 million articles from 13,900 journals available from the ArticleFirst and ECO (Electronic Collections Online) full-text files—a database that grows at over 750,000 records and 1,000 titles each year. The enhanced file will also cover Internet/Web resources.

Jay Jordan, OCLC's CEO (who will be interviewed by Tom Hogan in the December 2000 issue of Information Today), certainly believes in the concept, particularly as it serves to expand relationships with and for libraries:

OCLC today serves 36,000 libraries in 76 countries, and you can say that's a very impressive number. I quite agree. But, I am trying to look forward. Why isn't that a 136,000? Why not 1,036,000? And I don't limit our vision to just libraries. We should be linking every knowledge object that resides in every knowledge repository—obviously archives and museums, but also historical societies and local collections and library collections, most importantly….

Choice to the knowledge seeker, that's what we should be about. In our case via the library portal … We'd love to have an intelligent conversation with Amazon. We'd love to have an intelligent conversation with Barnes & Noble and Borders and so on. We need to look for partners that are interested in cooperating. Why would they? Access to 36,000 institutions around the world—and that's just the institution count. Do you want to tell me how many patrons are actually using those institutions to research? The point is there is a tremendous reach, and there are a lot of other players out there. We think there will be enough interest…. It's about intelligent partnering by OCLC and providing greater choice or [a] richer experience ultimately to the patron.

All in all, it looks like the big push toward a true virtual library service across the Web has begun. Win, lose, or draw, this is the big one.