3) Taylor: Question negotiation and information seeking in libraries.

In "Question negotiation and information seeking in libraries,” Taylor summarized and described the ways that librarians interact with their patrons to help uncover and solve their information need. While not an empirical study, Taylor's article was a valuable resource: his four levels of information need and five filters are attempted to explain the rather nebulous ways that a user/patron's information need was discovered and solved.

The librarians that Taylor interviewed for his study were described as "special librarians and information specialists" (Taylor 1968, pg. 29). These people were dealing with users/patrons who, Taylor suggested, were at a higher level of education and intelligence than a typical library patron. Even with this advantage, the librarians found that often people weren't sure what they wanted - And even when they did, it was difficult to articulate. I think that this is where a librarian or information specialist shines; they can comprehend and do things that are impossible for a computer or system.  Through the five filters Taylor described how a librarian was able to extract the patron's true need.  A computer, as advanced as search engines and systems are currently, can only anticipate a user's need to a limited degree and the AI as it is now doesn't have the knowledge or experience of a human with a library background.  By determining the subject (filter #1), understanding the motivation and characteristics of the patron (filter #2 and #3), the search system itself and its strengths and limitations in terms to the query (filter #4), and what sort of answer the patron will deem acceptable (filter #5), the librarian can guide the search (Taylor 1968, pg. 31-32). These five filters are not without flaws, however; human error can affect every stage of the search.  In the first filter, the librarian runs the risk of misinterpreting the user's desired topic or not discovering it at all. In the second and third, the librarian's biases may come into effect when s/he attempt to judge what the user's motivation and background are. In the fourth, the librarian or the search system can misinterpret the query. Finally in the fifth, the librarian can give the user an incorrect or incomplete search result.

I found Taylor's four levels of information need (Taylor 1968, pg. 32) very interesting. The first stage is the visceral need; the user might be unconscious of his or her true need at this point. In the second level, the conscious need, the query exists in some form in the user's mind, although s/he might not have a clear or firm searchable question. The third level, which Taylor calls the formalized need, has a developed query; the user has a distinct question which can be inputted into an information retrieval system. By the fourth level, or the compromised need, the user has taken the limits of the search system into consideration when phrasing his/her question. At this point, the original query has traveled from a vague idea into a fully formed, articulate question that can inputted into a system with the anticipation of retrieving results.

Taylor's article is as important research for me as a potential librarian because it portrays an information need as a process, and as a process that needs a librarian to guide it.