“The human natural tendency in information seeking is to fallback on passive and sampling and selecting behaviors derived from millions of years of [evolution]” [1]

–Marcia J. Bates

Today, the relationship between information retrieval (IR) and its human counterpart is seen as both important and necessary in any realistic discussion of information-retrieval models. In “The design of browsing and berry-picking techniques for online search interface[s]”, Bates proposes a shift from the machine-like classic information retrieval model to a more organic, user focused retrieval model. In 1989, when Marcia Bates wrote her article on “Berrypicking…” a user-centered searching model was a revolutionary idea. Prior to the induction of Bates’ model, IR models were often systems-centered rather than user-centered. If one takes this user-centered concept and applies it on a wider scale it makes even more sense. Society in general has been moving toward creating computer models that function more like humans. Take, for example, the military computer WOPR in the 1983 film “Wargames”. In this film, the major character matches wits with a computer gone awry, this computer has been built to emulate human-like thought processes and therefore reacts in a more organic and changeable manner to the main characters (the users) actions. This is exactly the type of IR model that Bates suggested six years later with her “berrypicking” concept. Bates suggested a more organic, ongoing, transformative process of answering queries in response to the users input, rather than the rigid, linear process of the classic IR model.

The essay is divided into five sections: “Introduction”, “A Berrypicking model…”, “How and where users search for information…”, “Search capabilities for a berrypicking search interface” and the “Conclusion”. Most sections are accompanied by useful graphs and examples that help to relay the information that Bates is presenting to the reader. She is direct and to the point and uses very colloquial language that supports her purpose of IR being discussed not only in the realm of scientists, but also in the realm of library professionals. The largest portion of the essay is devoted to the explanation of “six methods of search” and how these search methods should be incorporated into information retrieval systems. They are used by Bates to illustrate how search patterns vary organically rather than strictly following a linear pattern. The six main search strategies that Bates discusses are: “footnote chasing” (looking to the footnotes of a paper for more sources), “citation searching” (looking for other papers that cite from the same source), “journal run” (methodical searching through each journal for the intended topic), “area scanning” (physically searching items/documents in the vicinity of the original document of interest), “subject searches” (scanning subject indexes for relevant material) and “author searching” (finding additional works by the author on the same topic). Bates also goes forward to state that these methods are not used singularly but often are used in conjunction with one another. Discussion of these types of search methods supports the main idea she presents in her article, that IR models should be more organic and less linear and that users generally do not find just one type of searching method sufficient to meet their information needs.

Essentially, the essay focuses on comparisons of her “berrypicking” search model against the traditional, linear IR model. Her model essentially differs from the traditional model by stating that “the nature of the query is an evolving one” and “the nature of the search process…follows a berrypicking pattern, instead of resulting in a single set” (p.6). Bates’ model proposed in 1989 is still very relevant to society in general and even more specifically to Library Science. Elements of the concepts introduced in Bates article are represented today in various internet and computer applications. For example, del.ici.ous (http://del.icio.us), a popular internet based application, attempts to link users with like interests together, therefore allowing the user to browse others users’ favorite links as a kind-of relational search method. In the library setting, Bates’ model helps put into context the types of searching capabilities that users might apply to any given IR system and thus helping developers build better systems. Bates’ concept of “berrypicking” also has strong and weak points; strong in the sense that it is malleable and can conform to different types of search methods in many different types of settings (internet, physical, electronic) and weak in the sense that it is a very difficult concept to apply systematically as its very nature is inconsistent. Overall, Bates presents us with a very though-provoking concept; many results of this can be seen in various IR systems today. In the end, Bates leaves us with something to think about. How can we, as library science professionals, provide our users with the most interactive and organic searching methods available?

[1] Moreville, P. (2005). Ambient Findability. Sebastopol, CA: O’Reilly Media, Inc. (p.61)