- From: Jason White <jasonw@ariel.ucs.unimelb.edu.au>
- Date: Thu, 2 Aug 2001 12:36:22 +1000 (EST)
- To: Charles McCathieNevile <charles@w3.org>
- cc: Kynn Bartlett <kynn@reef.com>, Emmanuelle Gutiérrez y Restrepo <emmanuelle@teleline.es>, WAI Cross-group list <wai-xtech@w3.org>
Actually, my own proposal is too broad: any XML document instance would satisfy my definition. There would need to be a further qualification along the following lines: the symbols must be primarily intended to be perceived and understood by human beings. With this qualification, an SVG document, for instance, wouldn't be regarded as text because, although it is expressed in terms of symbols drawn from a finite alphabet (namely those defined in the DTD), and in fact satisfies a grammar, the symbols aren't meant to be read and understood directly, but rather their function is to control the rendering process (and also to convey certain semantics). I am thinking here of the shapes and paths rather than of the text content of an SVG document. Perhaps the following set of definitions will work: Text content (my proposal with suitable emendations). Non-text content: information which does not satisfy the definition of text content, but which is meant to be perceived and interpreted primarily by a human being (sounds, graphics, etc.). Content (simpliciter): the union of text content and non-text content (that is, anything which is either text content or non-text content).
Received on Wednesday, 1 August 2001 22:36:26 UTC