by Cy Dillon and Lyn Gardner
As many of our readers know, Virginia Libraries is archived by the Virginia Tech Digital Libraries and Archives (http://scholar.lib.vt.edu/). This collection of previous issues, going back to 1996, is a valuable research tool as well as a historical resource for the Virginia Library Association. It also allows our writers and subjects to link directly to articles by or about them. In many cases, our articles show up in the results of a Google search. For instance, when I ran a Google search on the poet Jeffery Beam, the first result was his website, but the second was his Virginia Libraries interview from Volume 50, Issue 2.
Of course, looking through everything on the web is not an efficient way of searching past issues, and neither is going through all the tables of contents. Recognizing the need to search the archived journals and other materials in their collection, the Digital Libraries and Archives staff have provided the Verity Ultraseek search engine to their users since the late 1990s. This resource is located at http://scholar.lib.vt.edu/ejournals/ej-search.html and is managed by Lance French, Systems Administrator for Tech’s DLA.
French explains that this software was chosen because it was – and is – a popular commercial search engine used by many large corporate intranets. While it is not inexpensive, it is much better than a homegrown search function, and it is periodically updated. The product is reviewed regularly, and it is always possible that DLA will move to another search engine. Still, French is certain that reliable search software will always be a part of the DLA site.The educational potential of interactive sites … should not fall victim to political posturing.
My experience with Ultraseek suggests that it has the typical strengths and weaknesses of most keyword searching. It almost always retrieves what you want, but there is almost always much that you do not want. That is not a problem for experienced searchers, so I am confident that Virginia Libraries readers can find past articles quickly and easily.
All this comes up because the Directory of Open Access Journals (http://www.doaj.org/) has offered us the option of providing article level metadata for past issues of Virginia Libraries. The coeditors have considered recruiting volunteers to be trained to tackle this large and potentially productive project, but at this point we think that the existing search function at DLA is sufficient for most research needs. We would appreciate hearing how other VLA members feel about the situation, and are very much open to changing our minds. Our email addresses are listed on the magazine’s contents page, so feel free to comment.
Speaking of comments, how can we be expected to avoid a bit of a rant about the latest symbolically appealing but totally misguided attempt of our Congress to tame the Internet? In this election year, the House of Representatives has overwhelmingly passed H.R. 5319, which would force public libraries to block all social networking sites in order to protect children from sexual predators. The American Library Association’s response that education about the dangers of the Internet is much more effective than barriers to access is right on target. The educational potential of interactive sites – and the bill covers almost anything interactive – should not fall victim to political posturing. As the bill is written, it might even apply to library sites with reference chat, email links, or customizable pages for individual users. That is simply an indication of our lawmakers’ relative innocence about the Internet.
American children, on the other hand, know the medium all too well. It is estimated that half of America’s teenagers already belong to social networking sites such as MySpace, and restricting their access in the library will do little to curb the popularity of these free services. In fact, it may keep youngsters away from safe, supervised library computer rooms and push them toward going online when they are at home alone or unsupervised at a friend’s house. As a college librarian, I have seen the impact educational programs have on our students’ use of Face-Book. There is no reason to wait until young people go to college to address safe and intelligent Internet use. Why doesn’t Congress fund Internet education in public schools and public libraries? The dollars invested will yield far more results than symbolic prohibition of interactive websites.
Finally, speaking of the changes we face in this complex digital age, and returning to the subject of keyword searching versus cataloging, it would be impossible to ignore a topic that has fired up the cataloging community over the past few months: the decision of the Library of Congress to cease providing authority control for most series beginning on June 1, 2006 (http://www.loc.gov/catdir/cpso/series.html). Debate has raged on both sides of this issue. Digital proponents such as Karen Calhoun, in her March 17, 2006, report for the Library of Congress (“The Changing Nature of the Catalog and Its Integration with Other Discover Tools,” http://www.loc.gov/catdir/calhoun-report-final.pdf), have criticized current library catalogs as providing an outmoded structure that functions more as a barrier to than enabler of access. On the other hand, cataloging proponents such as Thomas Mann, in his reports on behalf of the Library of Congress Professional Guild (such as his reaction to the Calhoun report, “A Critical Review,” http://guild2910.org/AFSCMECalhounReviewREV.pdf, or “What Is Going on at the Library of Congress,” http://guild2910.org/AFSCMEWhatIsGoingOn.pdf), warn that digital technology is not yet developed enough to adequately take over for the traditional tasks of cataloging – and that scholars and others seeking more detailed or higher quality information will miss out on many sources not retrieved through keyword searches. Indeed, as a result of this debate, there are numerous recent examples of how keyword searching fails to adequately retrieve (let alone collocate) series records that may be found in the archives of cataloging listservs such as PCCLIST (http://www.loc.gov/catdir/pcc/pcclist-info.html), AUTOCAT (http://ublib.buffalo.edu/libraries/units/cts/autocat/), and OCLC-CAT (http://listserv.oclc.org/archives/oclc-cat.html).
While PCC and OCLC have stepped up to the plate to ensure that series authority work will continue, it is likely that more changes are in the offing. There are fears that other aspects of Calhoun’s report might lead to the elimination of Library of Congress Subject Headings. Despite the increasing power of search engines, the benefits of keyword searching, and the difficulty involved in manipulating controlled vocabulary subject headings in a library catalog, it remains true that given today’s technology, subject searches can still yield more accurate and textured results than a Google search. The problem is convincing users that it is worth the time to learn to use these subject headings – and in this digital age, where some information is always a click away, that is a big problem.
What will the future bring? True artificial intelligence that is capable of analyzing a book’s subjects and collocating similar works despite differences in terminology? Intelligent books that “listen” for relevant queries and step forward to announce their benefits to readers through a “smart” interface that would take over where modern search engines fail? At this point, we simply can’t say. For the present, however, perhaps it would be wise to pursue both paths, rather than ignoring or abandoning either. Collaboration could help us achieve one of our greatest goals – to provide maximum access for all kinds of users. Let’s not abandon time-tested knowledge about how to organize and retrieve bibliographic information; but neither should we ignore the promise of digital access to worldwide knowledge, just because it hasn’t fully materialized yet. We need to strive together to build upon our knowledge to improve our services – but we should not abandon what works before a replacement strategy has been developed that will truly provide an improvement. The ultimate goal might combine the assets of both views: to simplify the search process, yet achieve a higher rate of accuracy. To do that, until such time as we develop computers truly capable of subject analysis, catalogers are needed more than ever. Let the digital visionaries work with the catalogers to construct an interface that provides the user with desired searching ease while retrieving relevant and complete results through an intelligent but invisible query of detailed cataloging data “behind the scenes.”