Introducing a topic map module for Drupal

SESSION OVERVIEW
The Drupal topic map module brings many new and powerful features to categorizing content. In the topic maps data model, a topic map is built from players that play roles in typed association. For example, Homer Simpson might play the husband role, and Marge the wife role, in an association of type marriage, in a topic map called The Simpsons. Roles, players, associations, and associations are all topics. In this Drupal implementation, topics are embedded in posts in a simple, wiki-like syntax. On submission, a topic map filter processes that syntax, and generates Drupal content in a data-driven way. For example, this test association, when embedded in content in the

[test:maptest:My Cool Map[maprole1:mapplayer1] and [maprole_2:1600 Pennsylvania Ave. NW Washington, DC=geocode]]

produces the following page:

Moving top to bottom, we can see that:

1. A table of contents has been automatically generated for the post. (Notice how the table headings conform to the data model; all the rows in the TOC with the same ID under “Go” constitute a single assertion.)

2. Each cell is automatically generated to link to all other instances of the topic. For example, clicking on the “1600….” link will take you to all other instances of that topic, gathered together on one page (also automatically created, if it does not exist)

3. Given syntax like [maprole2:1600 Pennsylvania Ave. NW Washington, DC=geocode]], the module knows from the “geocode” keyword to use the geocode plugin to geocode the address, feed it to GoogleMaps, and return the map.

Because the topic maps module permits plugins, data-driven content can be created from topics in any association. The following screen gives more examples:

Here we see, again, data in the wiki-like associations driving the creation of links and icons — here to Bible citations, the Federalist Society, et cetera.
Again, the automatically generated table of contents enables access to all instances of a given topic; for example, a page to all references to Federalist 47.

Here is a screen shot that sums up the design and the functionality of the topic map model. Use cases are the blogosphere, DIY journalism, social software, and scholarship — anywhere that the graph-like topic map data model applies and high precision/high recall search and navigation are required.

TRACK
Site Building

AGENDA
I will step through creating all the above functionality on a real site, on a case by case, post by post basis. Where warranted, I’ll examine the source code — and maybe even ask any wizards in the audience for help making this better!

GOALS
I would like people to come away from the presentation excited by the topic map paradigm, and full of ideas for how to apply it in their own lives.

RESOURCES
More information on the module will be available at