What's in a topic map? (1/2) - exploring XML
What's in a topic map?
In our last extension of the XMLMap we closed with a discussion of ontologies and topic maps, both concepts for expressing semantics of resources. I received many questions on the rather abstract nature of these concepts, as well as their practical applications and implications. This installment tries to explain topic maps with examples taken from our daily lives.
Topics maps revolve around three basic concepts:
Topic Map Concepts
To illustrate these concepts, let's look at something familiar to all of us: a traditional back-of-the-book index, in this case my review of Practical XML for the Web. Under W3C we find (slightly modified to better illustrate the concepts):
W3C, see World Wide Web Consortium
World Wide Web Consortium,
see also DOM; SVG; XLink.
DOM specification, 97-98, 232
Mozilla-based browsers and, 106
What's in an Index?
Various features can be identified in this example:
- Typographical conventions are used to distinguish between different types of topics: XML standards are underlined.
- Similarly, typographical conventions are used to distinguish between different types of occurrences: references to the term's definition are shown in bold.
- Use of see references handles synonyms by allowing multiple entry points with different names to the same topic, in our case the acronym W3C.
- See also references point to associated topics.
- Sub-entries provide an alternative mechanism for pointing out associations between different topics, such as special cases, e.g. browser-specifics.
There could also be different indexes for different concepts, such as XML standards, Web browser software and their versions and the like. Furthermore, occurrences could be distinguished as various types such as definition, quote, reference, etc.
The key features of a typical index are: topics, identified by their names, of which there may be more than one, associations between topics ("see also"), and occurrences of topics, pointed to via locators, such as a page number.
Glossary and Thesauri
Other types of indexes are glossary and thesauri. A glossary is an index with only one type of occurrence, namely the definition, which is given inline, not with a locator. A glossary entry could look like:
XML: A format for structured data.
A thesaurus, on the other hand, emphasizes other aspects of an index. The key features of a thesaurus are the relationships or associations between terms. Given a particular term, a thesaurus will indicate which terms have the same meaning, which terms denote a broader or narrower category of the same kind, and which are related in some other way. To continue our example, a thesaurus entry might appear as follows:
- A format for structured data
- Broader terms
- Markup Language, structured data format
- Narrower terms
- XSLT, XLink, XHTML, ...
- Related terms
- SGML, HTML
The special thing about associations in a thesaurus, as compared to associations found in a typical index or glossary, is that they are typed. This is important because it means that the two terms are related and also illustrates how and why they are related. It also means that terms with the same association can be grouped together, making navigation much easier. Commonly used association types like "broader term", "narrower term", "used for" and "related term" are defined in standards for thesauri such as Z39.19, ISO 5964 and ISO 2788.
A related approach to representing the structure of data are semantic networks, usually depicted as conceptual graphs. This article could be captured this way:
[Michael] (article) -> [Column77]
(Square brackets denote concepts, and parentheses describe relations.)
By adding the topic/occurrence axis to the topic/association model, topic maps provide a means of unifying knowledge representation and information management.With this newfound clarity, our next step is to define topic maps through three key concepts (can you remember them by now?).
Produced by Michael Claßen
Created: Mar 17, 2003
Revised: Mar 17, 2003