Introduction

The World Wide Web (WWW) has developed rapidly in the past five years as an accessible medium for the distribution of information and entertainment to those who have access to the global Internet network. Using free or low-cost browser software such as Netscape Navigator or Microsoft Internet Explorer, which have been developed to run on a variety of desktop computer platforms, users can retrieve and display documents from remote computer servers or from local storage devices such as CD-ROM, which have been laid out in standardized HyperText Markup Language (HTML) and hyperlinked to other marked-up documents. HTML allows the inclusion within the document of graphical images and other 'multimedia' features such as sound and video files may also be incorporated, although these require additional software to replay them. In addition to displaying text and images, the potential of static HTML documents has been further enhanced by a number of recent developments which allow user-controlled active content :

i) PluginsSmall compiled applications which can parse specific file types
defined by the MIME type filename extension and display the contents within
the HTML document^1,2. Machine dependent and must be installed locally by
the user.
ii) Java AppletsSmall applications written in the Java programming language
and downloaded from a remote server as machine-independent compiled 'byte
code' when requested by the local client. This byte-code must be interpreted
by a local Java interpreter to create the executable machine code. They are
device independent, and tend therefore not to be optimised for system
performance.
iii) JavaScript An event-driven scripting language³, incorporated within an
HTML document, which allows dynamic user control of standard HTML page
elements and inter-frame communication.

There has been a rapid expansion of chemical applications which use Web technology⁴and standards have been introduced to describe the chemical information used^1,2. The impact of the Web upon computational chemistry software developers has already been significant. Within the last 18 months a number of developers have begun to target the corporate 'Intranet' market by providing modelling tools - structure drawing, query handling, data presentation & analysis - with a consistent Web-based interface which can be easily learned by 'web-aware' researchers and chemists^4-6. Such tools, e.g. Discovery.Net from Tripos, WebLab from Molecular Simulations, ChemScape Chime from MDL, InteractiveLab from ACD Labs, have applet- or plugin-based applications which run locally on standard browers and are connected to networked computational or company database servers. Similarly, Web-based interfaces have also been adopted by groups designing Laboratory Information Systems (LIMS), e.g.an in-house spectral database & structure-query system at Kodak⁷, and the U.S. National Institute of Standards and Technology has instituted the NIST Chemistry WebBook as an on-line resource of published infra-red and mass spectral data⁸. The rationale behind these developments is twofold. Firstly, they provide platform- independent applications, running on Mac, PC or Unix operating systems, which can be