SGML documents : where does quality go?

Abstract

Nowadays the broadcasting of information through the Internet-based World Wide Web (WWW) is simple, fast and attractive. But it is also easy and fast to transmit this information with wrong data, or in such a way that the consumer acquires a wrong knowledge (the only ensurance about data correctness is the human author). The more distant is the service provider (the person who produces the WWW pages) from the information owner (the source of data) the more accute is this problem.

In this context, it is usually required quality in the information dissemination process. But, actually, it is easier to state that than to understand deeply its meaning, and moreover than to assure error-safe document processing.

In this paper we intend to discuss the concept of quality in the production of documents and its relationships with the different actors in the process, aiming at the proposal of therapeutics to avoid some kind of semantic erros. Namely, we will present some extension to the SGML document type definitions (DTD's) that will enable the inclusion of some semantic constraints useful to guarentee the information quality.