Abstract

This paper discusses the techniques and the hardware and software required to make large quantities of images of moderate quality transparently available to the Internet, and does so from the point of view of an art historian who needs at least 30,000 images network-available for teaching and research, and currently has 12,000 in that state. Areas covered include the pros and cons of various kinds of input and storage, and the automatic generation from data records of Web pages, with or without inline images. The continuing utility of the laserdisk, especially when linked to a computer, is emphasised. Strategies for dealing with up to 1 × 10⁶ higher-quality images derived from a digital camera and stored on a ten-terabyte Tape Robot and making them to some extent 'future proof' in a Web networked environment are then described. An appendix offers a description of the ArtServe Web server at the Australian National University (http://rubens.anu.edu.au) and the material it contains.