The last question has a clear answer: access to Web of Science would greatly improve access to literatures and could substantially broaden the teaching of science. The problem is a very substantial cost for an excellent product for a small number of faculty and student users.Our students and faculty now access research and review literatures in the sciences primarily via online database searching through the Web, and the Science Librarian teaches searching skills in various courses. Frequently-used databases include Cambridge Scientific Abstracts, PsychInfo, PubMed, several FirstSearch databases, and GeoRef. In general, searchers use the Web interface to progress from keywords to descriptors, and retrieve abstracts and bibliographic information which leads to (a) our library periodical holdings, (b) InterLibrary Loan requests, or (c) (in some cases) online full text.
Some librarian-mediated searching is still done via Lexis/Nexis (paid by subscription --patents and medical materials for the most part) and DIALOG databases (paid by the search --for databases not available to end users). Among the databases accessed in the latter way are Chemical Abstracts, INSPEC, and SciSearch (Science Citation Index), all of which are very expensive at the standard rates (which include timed searching and record-download charges). DIALOG's "classroom instruction program" account can be used when the search is course-related, at a greatly reduced cost, but DIALOG's search interface is so complicated and arcane that end users would not find it satisfactory if they could access it.
Institute for Scientific Information (ISI)'s Science Citation Index is the most broadly useful of these DIALOG databases. In addition to the SciSearch files in DIALOG, ISI has versions of Science Citation Index in print, CD-ROM, and web forms, widely used in universities and colleges despite their horrific price tag. It is important to understand why this investment is made by our sister institutions:
Science Citation Index adds an important dimension to bibliographic searching because it allows the searcher to see a retrieved article in its intellectual context, linked both to antecedents (sources which it cites) AND subsequent work (sources which cite it). This is of enormous pedagogical significance, because students can quickly see that the enterprise of science really IS collaborative and interlinked, and can follow the development of scientific thought from article to article. Citation indexing is also important to researchers, who can use it to discover connections of which they were unaware by tracing an article's influences forward and backward in time. An even more daring (and exhausting) use of citation indexing is to look at clusters and networks of citation, following an idea or a subject over time and analyzing patterns to identify classic articles (those which many people cite) and review articles (those which cite many articles). Ingenuity and elbow grease along these lines can lead to remarkable empirical insights and discoveries in the history of science.Because ISI indexes article bibliographies from thousands of journals, Web of Science is an effective search engine in many scientific fields --though it is not as comprehensive as (and would not substitute for) Cambridge Scientific Abstracts, PubMed, or AGRICOLA.
Since relatively few Washington & Lee science students write honors theses, it could be argued that there is little demand for the sort of citation linkage unique to ISI's products. Most of the requests for citation searches that come to me originate with faculty (perhaps a dozen a year), and relatively few courses seem to require the sort of bibliographic research that would make use of the unique powers of citation studies. This situation might change if Web of Science was readily available, and if history and philosophy of science was more frequently included in courses.