About Aboutness

Introduction
Subject analysis is a vastly important and challenging area of information organization. The literature reviewed in this week's reading assignments has not only outlined the process of subject analysis but brought to the fore several difficult aspects that have kept information professionals unsatisfied with its current condition in general, and with indexing for information retrieval systems specifically. According to Taylor & Joudrey (T&J)(pg. 303), being able to identify “precisely an item’s subject matter (often referred to as aboutness . . .)” is of significant value. Of equal value, and generally less definitively accomplished is then carefully and accurately assigning appropriate terms in an index to represent that aboutness. In this reading response, a brief synthesis is provided of the M. J. Bates (not so brief) article and the Taylor & Joudrey text. They are supportive of one another in their discussions of aboutness and the difficulties in achieving a database and retrieval system design that is completely satisfactory to its users. The conclusion is that there have been many changes in the subject analysis space since Bates wrote her article, and there is yet more on the near horizon that will further solve some of the inherent challenges.

Synthesis of Concepts
T&J informs that much literature is extant that points out that aboutness is more than just the words in the text used by an author. Accurate aboutness takes into consideration, among other things, the intentions of both the author and the users of the resource. This suggests there must be a connection between the person developing the vocabulary of the index and the end user, or at least an understanding of the latter by the former. Otherwise, the index is of great value to the indexer, but less so to the user. Bates (pg. 4) reinforces this concept by pointing out that the information seeking user doesn’t know what the resource is yet. The user has no idea what it is about, or what it can answer, or even how to ask for it. The indexer doing the analysis has the record in hand. There is no gap in understanding the resource. The main challenge for the indexer is to add to that “hands-on” knowledge by anticipating how the user will search for the resource. Even to the point of anticipating which terms the searcher might use, based on the information need. If the user’s guesses in the search don’t match the indexer’s description, then the resource is not found in the search and it is of little value to the user.

Further, Bates contrasts the indexer’s experience (with the subject analysis process) to the information seeking experience of the user. The indexer is highly trained and educated in the subject analysis arena, whereas the user is not, and may not even be experienced in the search process. So, having their thought process match is unlikely, unless the indexer thinks in terms of the potential searcher.
The main points of the Bates article are that human, database and domain factors impact the subject analysis and search processes. Those impacts can be negative if indexers don’t follow careful and well thought-out methodologies. Inherently, the indexer wishes to build something of great intellectual and scholarly value. Potentially, the more logical and sophisticated the retrieval system design is, the less it will resemble the inexperienced searching techniques of the user. The indexer must take into account the user community as well as the domain or environment of the users.

T&J describes the process of analysis and assigning terms. The text gives three examples of approaches (or methodologies) to the process, which are identified by their authors, Langridge, Wilson and Lancaster (as a representative of the Use-based Approach). Each approach consists of steps that generally follow a basic outline of asking questions. These questions begin in a sort of general arena and move to more specific characteristics of the information resource being analyzed. For instance, “What is it?”; “What is it for?”; and “What is it about?” in Langridge’s approach. Wilson describes in his approach four methods, which similarly work from generalities to specifics. First (Purposive), “What was the author’s purpose?”; second (Figure-Ground), “Is there a central figure that stands out?”; third (Objective), counting references to determine which vastly outnumber the others; and fourth (Cohesion), “what holds the work together?”. This general to specific question-tree approach was likened by Bates (pg. 14) to the same way library users approach the reference desk. They start with a general opening question and, based on the librarian’s interviewing skills, the interview moves to specifics until the librarian has determined what the user really wants and provides a path to that information. It seems that the database searcher follows this same approach of trying general terms and getting more specific until the search has been narrowed to relevant and desired information.

The impetus for following methodologies such as those presented in the text and article under review here is to overcome the many challenges inherent in communication of information. People from different environments, cultures and background simply think of things differently. Anticipating those differences will aid in overcoming them. The suggestion by Bates to build a user-friendly front-end to attach to the logically developed database makes good sense and since the time of her writing, this suggestion has been implemented by most developers of retrieval systems.

In today’s Web 2.0 environment, retrieval systems are advancing to the stages of incorporating user generated aboutness, or tagging, which will take the process a step further in overcoming some of the challenges identified in the Bates article and the T&J text. According to Rolla (pg. 178), user tagging such as is present in LibraryThing has added richness beyond the library catalog. This is the positive result that Bates, in particular, and the T&J chapter suggest we must pursue in the information organization profession as we overcome inherent challenges of subject analysis.

Conclusion
Many studies have been conducted and much has been written about the subject analysis process (see Notes in T&J, Chapter 9, pp. 328-332) which has added to the discussion before and since the 1998 Bates article. These studies and the literature have helped to formulate new methodologies and to enhance existing approaches to subject analysis. As approaches and methodologies, like those considered in this week’s reading material, continue to develop with an eye on solving the aboutness challenges discussed, the search process will only improve.

References:

Taylor & Joudrey, Chapter 9, Subject analysis (pp. 303-332) and Appendix A. An approach to subject analysis (pp. 419-427)

Bates, M.J. (1998). Indexing and access for digital libraries and the Internet: human, database, and domain factors. Journal of the American Society for Information Science 49(13): 1185-1205.

Rolla, P. J. (2009). Can user-supplied data improve subject access to library collections? Library Resources & Technical Services 53(3): 174-184.