The modern library owns and makes available information in all kinds of media, from traditional printed materials to digital audio and video recordings, to online electronic resources. The  online library catalog provides the means to identify the  holdings of any given library. These catalogs are less intuitive than search engines, and library patrons often have trouble using them. The difficulties result from the greater sophistication of how catalogs are organized, outdated metadata standards, and especially poor displays of search results to the user.

The earliest library catalogs were bound books. They became obsolete as soon as the library acquired any new books. The librarians had to make new entries in them by hand, and after a while the catalogs became illegible. At that point, librarians could only write and print a new catalog.

At the end of the nineteenth century, libraries started to use a card catalog, which could handle new acquisitions simply by interfiling new cards. The card catalog likewise permitted separate cards for any number of authors, titles, and subjects. When the Library of Congress started selling sets of cards to other libraries early in the twentieth century, all of them had to operate under a single set of rules for cataloging and for filing cards.

Henriette Avram of the Library of Congress developed MARC, or MAchine Readable Cataloging in 1968; it became an international standard in 1973. Although its first function was printing catalog cards, it proved very useful for the development of online catalogs. MARC was one of the first metadata encoding standards. Metadata simply means data that describes other data. All library cataloging rules have therefore been a kind of metadata.

Most often, that term applies to what is necessary in order for one computer to exchange information with another. Although librarians were at the forefront of the development of metadata, they did not play a large role in the development of computer networks or the Internet. The discipline of information science has arisen to take care of that.

Modern catalogs have a number of strengths, derived from centuries of the development of rules for describing first books, then a rapidly expanding number of other media that may be found in a library. A library catalog is a database that contains a description of each separate title the library makes available. An online library catalog has separate indexes, not only for author, title, and subject, but also keywords, classification numbers, standard and non-standard publishers' numbers, and much else that earlier technologies could perhaps include, but not really use. For example, online catalog records can include the entire table of contents, a rich source of keywords.

Each separate record in the library database is the creation of an individual cataloger, who describes the book (or whatever else) according to rules that are themselves an international standard. The earliest cataloging rules were often badly organized, sometimes self-contradictory, and made sense only to other
librarians.

The current rules, called Anglo-American Cataloging Rules, Second Edition (AACR2), are based on a set of principles worked out over time at international meetings of librarians, with the intention that the resulting cataloging records should be intelligible to library users. Their basic structure is based on so-called International Standard Bibliographic Descriptions (ISBD), published between 1971 and 1977 to standardize the appearance of catalog cards. AACR2 is a metadata content standard. ISBD is the closest existing equivalent to a library metadata display standard

A library cataloging record combines exact transcription of information within the item itself (the title page and table of contents, for example), with what is called controlled vocabulary. Controlled vocabulary eliminates synonyms, so that only one official way of rendering a person's name, the title, and the subjects exists. The structure of the database is such that no one has to know these official forms in order to begin a search and that they are easy to identify and use. By using a combination of keywords and the controlled vocabulary, library users can generate a list of what the library has with little "noise," or information irrelevant to the search.

The antiquity of the MARC format has some unfortunate consequences. It came about at a time when rolls of paper tape contained all of the data. To this day, the computer must identify the correct record through information in the "leader" and then read the entire record in order to return results. Most modern databases are relational databases, and MARC does not work with the programming language used in that structure. Attempts to combine the librarian-friendly aspects of MARC with more computer-friendly XML or to replace MARC with a different encoding standard have not succeeded to anyone's satisfaction.

Vendors of online cataloging systems consult with librarians, but hire people from information science to design how the systems work. People with an information science background lack understanding of, and often respect for, centuries of library tradition and principles of library service. Each vendor's proprietary products  interpret MARC data differently, and no one offers a product that makes full use of all of the data that MARC records contain. Vendors allow client libraries to customize their catalogs to some extent.

As a consequence, it is no longer possible, as it was in the days of the card catalog, for patrons to use multiple libraries and expect all of their catalogs to work the same way. The library catalog is broken and likely to remain so. It is based on a content standard (AACR2) first issued in 1978, but continuously updated since then, and an encoding standard (MARC) and display standard (ISBD) that are even older. MARC has been superficially tweaked over the years. No online catalog to my knowledge uses ISBD at all. And yet the library world is currently developing a new content standard (the most satisfactory aspect of existing library metadata), arguing about what to do with MARC, and completely ignoring the utter lack of a display standard.

What, then, is a library user to do when confronted with today's dismal catalogs?

Do a keyword search first. If you know an author and title, the author's last name and a unique word from the title is a good start, but you can use  any other combination of good keywords. Unlike Google, most library catalogs require that you separate terms with "and," so if you're looking for Marley and Me by John Grogan,  "grogan and marley" (without the quotation marks) would be a good search.

Next, look at any of the records in the results list that look promising. You will notice hot links for the author and any other personal, corporate, or geographic name in the record and for all the subjects. These are the authorized forms. Take note of them. For example, the author of Marley and Me appears as "Grogan, John, 1957-." That distinguishes him from any other John Grogans, of which the Library of Congress authority file has six.

Some of the subject headings may consist of strings of terms separated by dashes. In a well designed catalog, it should be possible to click on any one term in the string; the farther to the left you click, the more general the search. For example, you may find "Dogs--United States--Biography--Humor." Clicking on "United States" will bring you to a screen that starts with "Dogs--United States" and shows you all of the available subheadings (such as Anecdotes, Pedigrees, Pictorial works, etc.) Unfortunately, not all catalogs provide that option; in these worse than usual catalogs, you can only select the entire string.

Not all libraries use the same subject list. Most academic libraries use Library of Congress Subject Headings; most public libraries use Sears List of Subject Headings. There are others. (All American libraries use the same forms of names.) That complication has nothing to do with computer technology and everything to do with what list is most appropriate for a particular library.

Search engines, unlike library catalogs, do not benefit from human intellectual activity making records and combining them systematically to make a unified catalog. They rely on algorithms to identify keywords to decide what sites to return and back-links to decide in which order to display them. Google has no equivalent of controlled vocabulary, although I does offer some disambiguation.

Online library catalogs are less intuitive than search engines both because they allow more powerful and sophisticated searches and because there are no national (let alone international) standards for how they should operate. Learning how to use them is worth the effort for the sake of the power of searching an indexed database and precision of the search results.

+++++++++++++++++++++++++++++++++++++

If you benefitted from this article or post please consider voting by clicking on the vote button at the top right and/or leave a comment.

To enjoy the writings of this author: http://www.xomba.com/user/allpurposeguru

Did you enjoy the post or article? Have you checked out Xomba? Would you like to be paid to post the same types of things, or other items? Get paid to write. Click the link below and sign up. It's FREE...no hidden expenses.
http://www.xomba.com/referral/777db842