Text as Data: Text as End Product

Texts do not necessarily consist of linear strings of characters from the ASCII-or the IBM's extended-character set. Three projects at the Norwegian Computing Centre for the Humanities are presented as examples of this. Character representation in computer memory, on VDUs and on printers is heavily device dependent. This is exemplified with references to existing computer equipment. This hardware dependency makes it necessary to use local character encoding to get the most out of equipment at hand. This again necessitates a standardized encoding to be used when textual data is exchanged. Such a standard should be developed within the SGML framework. The Norwegian Computing Centre for the Humanities sees the development of standards for encoding complex textual data as one of its main fields of interest in the forthcoming years.