Problems

I encountered several problems during the creation of this thesaurus. One major problem was the scope and exhaustivity of terms in the thesaurus. There are many words and concepts in the English language, and with the help of other existing thesauri, one could take the documents that were given and create a large-sized thesaurus based on terms indexed from the document. How would one know when to stop? This is a problem faced by many indexers and cataloguers. The negative aspect of having a huge thesaurus includes the longer time it would take to load/carry, either as an online item or as a book. Having the assistance of online thesauri such as the ASIS Thesaurus and the ERIC Thesaurus were useful for quickly discovering established relationships between terms. This greatly sped up the process of constructing the thesaurus because I did not have to create any hierarchical, associative or equivalence relationships from my head alone. However, the more terms one is faced with, the more difficult it is to manage the size of the thesaurus. I did not want to overload the thesaurus with terms that bear no relation to the realm of library and information sciences. A large thesaurus also meant more loading time. On the other hand, I did not want to underrepresent concepts found from the subject statements. Hence, trying to keep the thesaurus within a manageable scope was a difficult task. It would perhaps be easier if there were guidelines available on how many terms are expected in the thesaurus.

Term selection was the process that required a large amount of time and effort. To ensure that the thesaurus was not too large nor too small, I had to take extra care in what terms to select for the thesaurus. To this end, facet analysis was not an easy task. I had difficulty formulating facets that would best represent all the terms that were selecting to the database. Adding to this difficulty was the many terms suggested by the ASIS thesaurus. It was tempting to index all the terms that were listed, but I realized that it was impossible given the much smaller amount of this library budget. For example, should I pair the Research and Analysis facets differently because some people would consider "Research and Analysis" as one facet? Also, I was faced with the question “How many subject areas should I create?" I paired Communications and Technology together (not to be confused with communications technology) because the subject area appeared more appropriate for a Library and Information Science setting. One would expect the emphasis to be on the process of communication rather than on the area of technology. Other issues to consider during the term selection process include the use of singular and plural forms. Count nouns such as "libraries" and "archives" were expressed in plural form, while non-count nouns such as "medical research" were expressed in singular form. Hyphens were avoided except for terms such as "evidence-based librarianship," in which "evidence-based" is a combined term that describes the type of librarianship. Scope notes were used periodically to further clarify a thesaurus term. In dealing with compound terms, I made decisions based on whether the term would benefit more from retention or from factoring. For example, factoring of the term "World Wide Web" would hinder comprehension and thus should receive retention.

Facet analysis was difficult because the single-concept terms that I have selected had to correspond with the main facets.  I had no trouble coming up with a long list of terms that represented the different concepts associated with the original concepts expressed in the subject statements. Sometimes the association between terms would be obvious.  For instance, the identifiers "Britain," "Canada" and "North America" clearly indicated a geographical-oriented facet. At other times, associations could be vague, and it would be difficult to fit terms into the facets that one has created. For example, with "antiquities," I created the BT History.  Where would that fit in Information Sources, Science, Geography, Education, Population, and Communications?  Also, did the facets accurately sum up the isolates that were formed?  These questions were consistently raised during the construction of this thesaurus. It is important to always have essential questions in mind while one is in the process of constructing a thesaurus because one must not forget the needs of the end users. One should not be carried away by term selection and creation of relationships to the point of losing sight of what the terms in the thesaurus are meant to reflect. This is dangerous because one could wind up adding terms in the thesaurus that would not benefit the users and would instead inhibit the functionality of the thesaurus. With regard to this thesaurus, it is important to keep the context of library and information science users in mind, then select terms accordingly.