& 
.
This secondary text provides background and argument for an important point in our article about the use of hypertextual glossaries, specifically why and how we believe that glossaries should be considered as a form of annotation. There has been much interest in annotation of texts (both as traditional paper-based documents and in software systems) in recent years, most notably at the ACM's CHI and Hypertext conferences. Some of that research has focussed on implementation technologies with some attention paid to the human needs that will drive the acceptance of such technologies. Other research and speculation has concentrated on human needs and uses for annotation.
We believe that useful lessons can be adapted from the broader investigation and development of annotation-based technologies in the more limited area of glossaries.
Annotations are notes or other material that are added to published
works (such as books and webpages) by someone, or something, other than
the original author.
Annotations may appear at any location within a document, do not
necessarily include links, and are not restricted to any particular
form.
O'Hara & Sellen [1997] conducted a classic study of readers using both paper
and electronic tools.
The participants in that study indicated that annotating and
note-taking while readings was important in deepening their
comprehension of the text
[O'Hara & Sellen 1997, p. 337].
Campbell & Maglio [1999] showed that, at least some types of computer-based annotation can be helpful to users without also being distracting. XLibris [Price et al., 1998; Schilit et al., 1998] is perhaps the best known stand-alone computerized annotation system. It does not yet have any specific features supporting glossaries. Fishkin et al. [2000, pages 79 – 80] reported on development of computer-based interfaces that are more useful than traditional media supporting annotation tasks.
For reasons of space and focus we cannot provide a more comprehensive review of annotation here. We do however point interested readers to the articles by Ovsiannikov et al., [1999] (for a review of annotation technology as of 1999) and Marshall & Brush [2004] (for a study and more current, but briefer, review).
Our main interest in examining annotation here is to explain why
we believe that glossaries can be fruitfully considered as a
distinguished type of annotation.
Glossaries are structured presentations of (single
and multiple) words that refer to topic-specific meaningful concepts
(in the text or the glossary itself) and definitions and commentaries on
those words.
We refer to the (single and multiple) words that refer to
topic-specific meaningful concepts
as terms.
Glossaries are thus restricted to the general form of an
entry: a tuple in which one part is the term and the other the
text that is associated with the term.
The associated text may contain multiple definitions or comments, and
references to other entries in the glossary.
The associated text need not be structured in any particular way.
The example from §2 includes specific parts for two types of
cross-reference links: the q.v.
which is used, in the example, for reference to a super-class, and the
c.f.
which is used to point to
potentially synonymous terms.
Figure 2-I-1 is reproduced from Section 2 (Two Hypothetical Use Scenarios
).
We consider glossaries to be a specifically structured and limited form of annotation. Annotation can be a powerful tools for extending a document, but as with all powerful tools it has its risks. Because annotations in general are free-form, there is less opportunity for the technology to help its users to make intelligible notes. We are in not in any way seeking to detract from the use of annotation as an appropriate technology. We do however propose that glossaries should be studied as a potentially valuable tool to be used alongside annotation in general.
References for all works cited are available in a separate chunk.
& .