downloaded viaan Internet link to multiple users from a central source, allowing rapid and centralized software maintenance⁹. Secondly, it is hoped that the provision of standardized Web applications will both reduce the need for user retraining with new software developments and also promote better communication and information exchange within research groups.

Discussion

1. Standalone JavaScript Application

The initial proposal for VChemLab was to provide a single Web-based information resource that could be used in conjunction with practical laboratory teaching courses here at Imperial College. Information for such courses - molecular structures, physicochemical data, reference spectra, safety & toxicological information and practical details of synthetic procedures - is often poorly available, scattered around laboratories and libraries in manuals which are easily damaged or mislaid. The development of VChemLab would provide chemistry students, who are generally highly computer literate, with an accessible and intuitive computer-based source of such data, which could be readily updated to include new information or subsequent changes to the course content. VChemLab was therefore first developed as a standard client-server operation with no control or limitation placed over user access. Initial work concentrated on developing the structural organisation, and an early decision was made to employ JavaScript³ in conjunction with the standard HTML Web page documents in order to permit the dynamic control of the displayed information through client-side searching of JavaScript data arrays (Fig.1).

structure & spectra files

HTML/JavaScript pages

JavaScript arrays

data & image files

Figure 1. VChemLab client-server architecture

Information used within VChemLab may be divided into four types and this is initially stored in files under the operating system directory structure on the server (Table 1).