In this post, I will summarize the development of online library catalogs. Once I have pointed out the problems and the reasons for them, I will explain how you, the ordinary users, can make the most efficient use of them for finding the information (or entertainment) you want.
No perfect technology for library catalogs has ever existed. The earliest catalogs were bound books. Every time the library got something new, a record of it had to be written in the catalog. With the card catalog, the library could update its holdings continuously without ever having to disturb previous entries. For the first time it could also provide separate entries for title, author, and subject.
In the early 20th century, the Library of Congress began to sell sets of cards to libraries all over the country. That service required standardized rules, which in turn meant that the public could count on one card catalog being organized just like any other. It took a lot of space, though, and because it relied on humans to file the cards, it was inevitably full of mistakes despite librarians’ best efforts at quality control. Online library catalogs solved these problems and caused others.
The beginnings of online library catalogs
All library catalogs consist of data (cataloging records) about data (whatever is described in the record), and the modern word for that is “metadata.” Usually, metadata refers to one of the necessary preconditions for two or more computers to exchange information with each other. The library world achieved the first breakthrough in automating metadata when the Library of Congress’ Henriette Avram developed MAchine Readable Cataloging, or MARC in 1968. Although intended as an efficient means of printing catalog cards, it proved very useful for developing online library catalogs. MARC became an international standard in 1973
The field of information science has since emerged to develop large computer networks, including the Internet. It has developed modern metadata standards that combine standards for content (the information contained in a database), encoding (the programming language necessary to load that information into a computer), and display (what the computer’s output of database information looks like to the end user).
MARC is the earliest of encoding standards, invented when computers stored data on paper tape. Today’s computers must still read a MARC record from beginning to end in order to extract data. People trained in information science have since developed relational databases. They don’t like to use such a clunky and obsolete approach to database encoding as MARC. They, and not librarians, design modern catalogs. Each catalog vendor has its own proprietary way of dealing with MARC’s limitations, and not one of them uses all of the data contained in a MARC record.
The closest the library world has come to a display standard are the International Standard Bibliographic Descriptions (ISBD), published between 1971 and 1977. Their purpose was to standardize the layout and appearance of catalog cards. To my knowledge, no online library catalog displays information in ISBD. They all use various labeled displays. Because their approaches to encoding and display are proprietary, and because some of them give individual libraries a great deal of leeway in tweaking the system, it is no longer possible for the general public to learn to use the catalog at one library and expect to be able to use the catalogs effectively at any other library.
Library content standards have been developing for centuries. The current standard, Anglo-American Cataloging Rules, 2nd edition (AACR2), appeared in 1978. It has been subjected to constant, and sometimes drastic revision over the past thirty-odd years, but it is about to be replaced with a new content standard, called Resource Description and Access (RDA). It is even more international in scope than AACR2. Unfortunately, RDA remains strictly a content standard, with no attempt to incorporate new encoding and display standards to go with it. The Library of Congress, the National Library of Medicine, and the National Agricultural Library have jointly decided to implement RDA no sooner than 2013, subject to some remaining problems being solved.
Basically, then, the library world has spent years updating its content standard, which was already the most robust and up to date of the three kinds of standards. It has left the obsolete encoding standard and the non-existent display standard unexamined. To be sure, the MARC standard has also undergone continuous revision, but without any successful means of making it into a relational database. In my not so humble opinion, the public would have been better served if the international negotiations on a new content standard had been spent instead on encoding and display, which remain broken.
RDA will allow improved ways of describing works, expressions, manifestations, and items that a library obtains. Don’t worry if you don’t understand those last four bits of jargon. Librarians are still arguing over just what they mean. Without a new encoding or display standard, vendors of online library catalogs will probably interpret RDA data only partially. You probably won’t notice much different until your local catalog contains a large number of RDA records.
So how am I supposed to use an online library catalog?
Remember that a library catalog works on a database. You can’t approach results the same way you do with a search engine. For one thing, the resulting list will not directly include the information you want; it will give a list of resources that contain your keywords. For another, it stores information in separate indexes, which in turn allows for more focused searching.
If the catalog has a single search box that looks like a search engine search box, look for a link that says something like “advanced search.” You’ll find it much easier and more flexible. It will let you limit your search to the name, title, subject, or numeric indexes of the database. It won’t make you limit your search if you don’t know which index you want to concentrate on.
Start with a keyword search, using the most specific terms you can think of. In a general keyword search for example, you can look for a particular resource by combining the author’s last name and a word or two from the title. (You will probably have to insert “and” between terms, which is not necessary in a search engine.)
Your search will turn up a list of resources from the library’s collection that matches your search. Choose some that look promising and look at the records to see what the hotlinks are: the author and all other personal, corporate, or geographic names; all subject headings; and perhaps something called a uniform title in AACR2 and perhaps more helpfully, a preferred title in RDA. All of these hotlinked terms are authorized forms from an authority list, or controlled vocabulary. That is very important, because library catalogs have one unique form for every name, subject, or preferred title. Click on any one of them to find everything that the library has by or about the hotlinked term.
For example, the book Marley and me was written by John Grogan. The Library of Congress Name Authority File includes six authors named John Grogan. The one who wrote Marley and me appears as “Grogan, John, 1957- .” If you find a catalog that does not properly hotline all of the terms labeled “personal name,” for example, and want to follow up on one of them, simply copy and paste into the “author” search box to get a list of everything by or about that person.
A book (or play, or poem, or opera, etc.) written in a language other than English may have multiple English titles. In order to keep them all together, the library world (and in the US, specifically the Library of Congress) devises a preferred title that consists of the authorized form of the author’s name and his or her original title. A play by Bertholt Brecht known in English as The Caucasian chalk circle will appear in the authority file as “Brecht, Bertolt, 1898-1956. Kaukasische Krederkreis. English.” Why? That puts the original German, in this case, together with all of the translations into any language, including any English translations that may appear with a different English title.
Subject headings can be difficult. A book like Marley and me may have a subject heading that looks like “Dogs–United States–Biography–Humor.” The online library catalog should (but not necessarily will) have each of those four terms hotlinked separately. If so, you can click on any one of them. The farther to the right you click, the more specific the search. If you click on “Dogs” or “Dogs–United States” you will get a list of all available sub-headings under those more general headings.
Search engines have no equivalent of controlled vocabulary. Unless you choose specific keywords very carefully, you are likely to get a result set that does not match your intentions very well. Despite the lack of international display standards and an up-to-date encoding standard, the online library catalog enables more powerful and sophisticated searches. Even the ideal online library catalog may be, or at least seem, less intuitive than a search engine. On the other hand, it is worth whatever effort it takes to learn to use even the clunkiest catalog for the power of its indexing and the precision of its search results.