Representing Information

[Next] [Up] [Previous]
Next: Ordering the Possible Up: AsTeR -Towards Modality-Independent Electronic Previous: AsTeR -Audio Documents

Representing Information

  All information has structure, and any physical rendering of a document is a projection of this structure onto a particular medium, e.g.,printed paper. A ``rendering'' of a document on some medium is best understood if it makes this logical structure readily apparent. For example, a visual rendering -onto a two-dimensional medium like paper- may use cues like boldface, different fonts, and indenting to help reveal structure. A visual rendering takes advantage of the eye's ability to rapidly access different parts of a two-dimensional display. An audio rendering has to use an entirely different set of cues to reveal structure.

Early in the development of AsTeR , we realized that the ability to render information in a variety of output modalities would be a direct function of the richness of the internal representation used to capture structure and content. Abstractly speaking, the high-level structure of a document is independent of any particular mode of display, and the internal representation should reflect this. As a first step in realizing AsTeR , therefore, we developed high-level models to represent document structure. For instance, the richness of the representation used by AsTeR completely frees the order in which subterms in an equation are rendered aurally from the order in which they would appear on paper. (See 4 for details. )

This section briefly outlines some of the representations used in AsTeR . Rendering this high-level representation is outlined in s:rendering. Based on these ideas, we define a set of requirements in s:conclusion that should prevent electronic encodings from being tied down to any single display form.

TV Raman
Fri Mar 10 08:30:23 EST 1995