Tuesday, August 17, 2010

Browsing versus Keywording

A lament sometimes heard in the age of electronic research is that browsing is no longer possible as it once was, say, in library card catalogs. Fingering the rows of cards in the long wooden drawers, which might be devoted in one half of the catalog to authors and titles, and to the other half to subject topics, led to unexpected surprises. Many of these wooden catalogs were sold as collector's items, and now do indeed collect a variety of small things in private homes. Decades ago, the cards in the catalog were handlettered in a distinctive library script, which was taught in library schools. How sad to have lost even a simple but clear hand-lettering such as this. (Though several staff members in the Drew Library write a beautiful script.)

But in fact, the concept of browsing still exists in online library catalogs and periodical databases. A researcher browses, in this sense, when she locates herself in an ordered list of terms or numbers, in which she can move up or down. Anything so orderable is browsable: titles of books, names of authors, call numbers, subject topics. In catalogs and databases, the ordering is alphabetic in the case of author names, book titles, and subject topics, and numeric in the case of Dewey Decimal call numbers; it is alphanumeric in the case of scripture citations. For instance, if I browse-search in the Drew catalog for the title of a book, I find myself in a very long list of all the titles the library owns. So browsing can also be a way to take the full measure of a library or database, to see all the books it owns, or all the journals it indexes, or all the authors whose works it contains.

Most catalogs and databases offer a button for browse searching. If you click on that, you are primed to scan alphabetic or numeric lists. For example, if you are in Browse mode, and search for a subject term, you will be placed in the alphabetic list of subject terms that catalog or database uses to describe items, at just the point your search term occurs in the list. Or, if your term does not occur there, you'll be placed at what would be its alphabetic neighbor. From there, a single click on the term usually takes you to citations for whatever items (books or articles) are associated with that term in the catalog or database.

Browsing is useful when you aren't sure if a database recognizes your search term; or if you're unsure of a spelling (as in the case of author names). Browse searching can also give an overview of a topic, and show how it subdivides. This is especially useful when researching a broad topic like, say, Good and Evil. A browse-search for the subject, Good and Evil, within the Drew catalog shows it subdivided by: History of Doctrines, Psychological Aspects, and Religious Aspects. A browse-search can also locate a topic in a larger context. For example, browse searching the list of scripture citations in the ATLA Religion Database will retrieve articles that discuss your verse or pericope both by themselves and in larger contexts of verses. For example, if I browse-search for Genesis 1:1 in the ATLA scripture citation index, I learn that there are 30 articles on Gen 1:1, 7 articles on Gen 1:1-2, and 3 articles on Gen 1:1-10. Browse searching the subjects in the Drew library catalog shows me the complex way the library describes books about Genesis 1:1, like this: Bible O T Genesis I:1 (that's a Roman numeral for the chapter, and an Arabic numeral for the verse). The catalog shows me that there is one book on Gen I:1-3, but many commentaries and interpretations on chapter I as a whole, and still many more on larger sections of Genesis that include the first verse.

Keyword searching follows a different logic. In this case, I am simply matching a term I feed the database to terms already in the database. Keyword searches retrieve the records in the database that contain the search term I fed it. If I limit my keyword search by Author, I retrieve records that identify my search term as an author name; if I limit my keyword search by Subject, I retrieve records that identify my term as a subject topic. If I leave my search unlimited, I retrieve records that contain my term in any way at all, as author, title, subject, or in any other way (for example, as the name of a publisher, or the title of a chapter in the book).

The Drew catalog invites both browse and keyword searching. Keyword searching is the default mode. If you want to browse-search, click the Browse button at the top of the catalog. You can Browse-search Authors, Titles, Series Titles, Periodical Titles, and Subjects. If, from Browse-searching, you want to return to Keyword-searching, click the Keyword button along the top of the catalog. Within the Keyword mode, you can limit your search by Author, Title, or Subject, or, leave it unlimited, and search throughout all the records, by clicking on the second Keyword button that appears beneath the search box. It is important to distinguish between the two Keyword buttons. The one above the search box is in distinction from browse searching. The one below the search box presumes you are performing keyword searches, and simply executes an unlimited search (in distinction from a search limited by Author, Title, or Subject).

Be on the look-out for browsing options in the databases. ATLA allows many kinds of browsing--by author, title, journal title, subject, scripture citation, for instance. Access to these is via the Indexes button that appears along the top of the screen. You'll find a browse option in America: History and Life also via the Indexes button. The social science databases (PsycInfo and SocIndex) go one better than that, and offer a thesaurus of subject terms used in those indexes. In Soc Index, the button for that is Subject Terms; and in PsycInfo, it is Thesaurus.

1 comment:

  1. Excellent post. In the past, I'd always thought that browsing the stacks was important for making serendipitous discoveries - discoveries that electronic catalogs could not replicate. However, I was recently reminded of something obvious : many books that should be on the shelves may be checked out! So, the catalog and various electronic lists really reflect the entire holdings of an institution, not the shelves. There may be some serendipity involved to physical browsing, but I'm frankly a little less enamored of it than I used to be.