D5.1orig-Appendix G - General description of the Hulton Archive Components

From EChase
Jump to: navigation, search
Warning: this is a copy of the D5.1 document as originally submitted to the EC.

To contribute please use: General description of the Hulton Archive Components.

Back to D5.1orig System Specification

The original requirement, in 1994, for the Hulton Digital Archive (the Hulton Deutsch Library as it was then) was for a database system to support the digitizing program, support for captioning the digitized images, i.e. the addition of descriptive metadata, and finally a graphical user interface for searching and retrieval of the growing image repository for use by library customers, mainly picture researchers in London newspaper offices connecting via ISDN.

The database component

The major component provided was Index+. A record structure with up 40 fields was defined to describe the images. A file system was provided to receive the digitized images and thumbnail and display image sizes agreed. Processes were then implemented to create the required thumbnails and screen images from the reference digitized images. The resultant image file system, called the image arena was then linked to the Index+ caption database. The digitising process was carried out in a specially created light controlled environment. The system provided for the creation of CD-ROMs of the images as well as downloading to the image arena.

The captioning process

The Hulton Archive Library had existed since the 1940s as the major photojournalism archive in the UK and had a long history of service provision to newspapers, publishers and broadcasters. Captioning techniques had evolved during this early history in response to market demands so that by the time the possibility of computerizing the collection came along in the mid 1990s manual techniques of classification and description were well established and highly effective. Our goal was to carry these techniques over into the computer based archive so that researchers could carry on using the search terms and classifications they were familiar with.

The captioning process used external and internal captioners. Captioners were people familiar with the Hulton Archives classification schemes and able to describe the images using the required terms. External captioners were sent, by post, CD-ROMs with a subset of images and captioners then added the image captions in the field structure defined for the database. The set of captions was then sent back to Hulton on a floppy disc or discs which were uploaded into the main database and linked to the digitized images in the image arena. Internal captioners worked online with partitioned subsets of the image arena. This workflow functioned very well and within a very few years the digital archive had grown to over 100,000 images and associated captions.

The Index+ terminology system was configured with the Hulton search term classification thesaurus and this supported the basic keyword searches required.

The search interface

The third major component of the Hulton Archive system was the graphical user interface. In 1995 the intention was to provided online access to selected customers via ISDN. A user interface, designed by our subcontractors, Diverse Interactive, was implemented giving users a very intuitive, easy-to-use set of search and retrieve functions. Searches could be free-text, keywords or dates or some combination of these. A light box facility was provided so that users could build up a set of images for future purchase. When they had selected images they wanted to license they would phone the Hulton Archive sales department for pricing. The conventional business processes of invoicing and delivery of prints then took over.

Subsequently of course ISDN was rapidly overtaken by the internet and we then provided the Index+ Web Gateway for the Hulton Archive web development programme.

Back to D5.1orig System Specification