OCLC’s CORC in the Library – Cooperative Online Resource Catalog

OCLC’s CORC in the Library – Cooperative Online Resource Catalog – Brief Article

Terry Ballard

The Cooperative Online Resource Catalog is good news for this university

The whole e-book situation went beyond theory at our library at the end of the Fall 2000 semester. One night, I happened to check Qcat–Quinnipiac University’s OPAC–for Mark Twain’s Huckleberry Finn. QCat showed that the library owned two copies, and both of them were checked out until the last day of classes. Fortunately, we had put in a link to an online version of Huckleberry Finn, so a student who really needed it could start reading with the click of a mouse. Being online in 2001 means that you have potential access to a growing library that contains all of the world’s major classics prior to 1900.

It also adds a new level of complication for those of us who work with online catalogs. These machines have been adding more and more clever ways to answer the questions that the card catalog answered for our parents: What’s in the library, and where can I find it? Adding links to online resources challenges our concepts, because the library does not “own” them in the classic sense. The scanned pages of Huckleberry Finn are on a server at the University of Virginia. Do we provide MARC records because the library owns the books or because we can use the catalog to make them available to our users? If the latter is true, then it’s perfectly legitimate to include that electronic item in our online catalog.

If we’re going to make links to Web resources, then we need to go beyond the first tentative step that we followed in our library: adding a link to an existing MARC record. We should have a record that describes in detail which edition of the book was scanned, who was responsible for the work, and what the book is about. Fortunately, OCLC has been working on such a project, and all of us will be the winners.


The OCLC Cooperative Online Resource Catalog (CORC) began in January 1999 as an initiative to help librarians make sense of the free resources available on the Web. According to Bill Carney, a consulting market analyst at OCLC, the initial goal was to sign 100 libraries up in a test phase during which they would select Web resources and catalog them in standard MARC or Dublin Core format. Dublin Core was devised by OCLC specifically for describing electronic resources. By June 2000 the CORC records were added to the main OCLC database.

In the December 2000 issue of Information Today [page 26 and http://www.infotoday.com/it/dec00/hogan.htm], OCLC CEO Jay Jordan described the project to Tom Hogan: “What we are doing with CORC is involving libraries in a selection process and producing, on behalf of library patrons, a mechanism to screen out some of the garbage that we all know is out there on the Web. This is just simply an additional category of information objects from our standpoint. But we’ve done this on a global scale, and we have 489 libraries from 24 countries from around the world participating with us in a) developing the tool set on the fly and b) building the catalog of Web resources.” In other words, CORC catalogs the Web’s resources in the same way that OCLC libraries have treated other media–giving the sites full subject headings and call numbers. The ability to catalog CORC records has since been thrown open to any OCLC member institution. Librarians who want to investigate CORC need only go to its Web site at http://corc.oc lc.org and log in with their library’s regular password to get to the cataloging module.

After I logged onto the service, I searched for “Quinnipiac.” It turned out that seven Web pages had already been cataloged, but all of them bore the name “Quinnipiac College”–a title that was discarded in favor of Quinnipiac University 6 months earlier. I was happily surprised to find that I had the right to make the corrections on the spot. Back in my cataloging days, only a few selected institutions had the right to correct a record in OCLC, so I approved of this democratization. It took an extra day or so, but later that week the corrected records showed up in FirstSearch’s WorldCat database.

The catalog wasn’t perfect. Most notably, there was a record for our library that was created before the institution’s Web page was redesigned, so it led to a dead link. When you call up a CORC record for editing, it displays the MARC record in the upper frame and the Web resource itself in the lower. After a half hour of trying, I could not correct the record, so I “cloned” it and made a corrected record from the copy. I passed the old record to our cataloger to see if she could discard it using the standard OCLC Passport software. Aside from this glitch, I found the user interface particularly intuitive. A dropdown box at the top of the screen allows you to edit, reformat, save the record with local corrections, or make a correction in the full database. Although the interface is very user-friendly, the first day I tried it, this process was so slow that I wasn’t sure if anything had happened. On subsequent days, it did its work in seconds, and the record I was unsure of was in FirstSearch the next morning . Another glitch was the record for QCat. When I called it up in CORC, an error message at the bottom said that the URL was invalid, but when I clicked on the link in the 856 field, it worked perfectly.

When using the option to catalog an entirely new record, CORC will not only display the resource, but extract data from the Header and Metafield data to create the record, and it will create a fixed field file for computer files. I found this to be quite useful, and it allowed me to create a new record describing our library’s Web server in about 15 minutes. The rest of the time was spent adding subject headings and descriptive material in the notes fields. An added bonus is the credit that our library gets for original cataloging. Later, I looked at the same record in Dublin Core format. At first glance, the new format didn’t look radically different from MARC.

I asked Bill Carney about the relationship between CORC, NetFirst, and FirstSearch. He said that NetFirst is still a very active agency at OCLC, in which the company’s own personnel select and catalog Internet resources. Records in NetFirst are automatically available in CORC, but CORC records do not display in NetFirst. In any case, I find it exciting that librarians will be working together to create a catalog of useful Web resources.

What to Do with It

Our library is always actively pursuing ways to increase access to worthwhile full-text materials on the Web. We already have a page of links to e-text sources at http://invictus.quinnipiac.edu/etexts.html, and users can look through those links to find particular works. However, our goal would be to use QCat’s indexing capabilities, so the marriage of MARC and Web resources is good news for us. I would like to add some of the more impressive scans–such as the ones at University of Virginia–to our catalog, and CORC is providing that capability. This will be a way for our catalog to add classic works that we may have missed in our paper acquisitions. It will also be a way to supplement a collection of e-books that we have begun to acquire through our subscription to netLibrary.

There are oddities with the concept of mixing books, media, and Web resources in one online catalog. One of the initial CORC volunteer libraries cataloged QCat. If I downloaded the MARC record that describes QCat into QCat, the catalog would have a record for itself, which would then provide a link to QCat. This kind of recursive situation reminds me of the puzzler that comedian Steven Wright posed: “What’s another word for thesaurus?”

Terry Ballard is the automation librarian at Quinnipiac University in Hamden, Connecticut.

COPYRIGHT 2001 Information Today, Inc.

COPYRIGHT 2001 Gale Group