CMS 2010: IBM Information Architecture Workbench Presentation

Day two of the Content Management Strategies Conference 2010 in Santa Clara.  This morning I saw an interesting demo presentation by Kristen Eberlein (formerly of IBM, and now working as an independent consultant), who was showing off the capabilities of the most recent version of IBM Information Architecture Workbench. This is the program that used by known as IBM Task Modeler, and in its current incarnation it is a free program for the Eclipse platform designed to allow users to quickly create a map and its associated topics.

She walked the process of creating a map, and then quickly creating a bunch of empty, “stub” topics that flowed from it. One of the purposes of the tool is to enable a Document Architect to create a draft of a map and the associated topics that are to go with it, which can then be handed off to the writer to fill in with content. It seems to be a good tool for rapid prototyping of DITA projects: just run the map and stubs through the output generator and then hand it over to the stakeholders for review. She also said that she found the program to be particularly useful when training people on DITA, as people can easily understand the information it displays; as she said “pretty much everybody understands post-it notes and trees”.

She said repeatedly that the current incarnation “is buggy”, and mentioning that earlier she had created a DITA 1.2 specification map with sample content, and that when she opened it up the program threw a Java exception that ended up erasing all of her data. Needless to say, she was not pleased.

She listed the following as some of its advantages:

  • good for speedy prototyping
  • good for working with stakeholders in a way that is easy for non-DITA users to understand
  • fast way to create stub files
  • can easily export screen shots of the map
  • can also export DITA map data to Excel
She also mentioned some of the following disadvantages of the program:
  • doesn’t currently understand topicref format=”ditamap”
  • adds many IBM-centric topicmeta tags to the DITA map
  • adds @type attribute to topicref tags
  • has bugs, still very much under development
  • support for DITA 1.2 is minimal
  • interface is increasingly busy with all the new specialized topic types
A copy of her presentation can be found at the bottom of this page.


"DITAWriter" is Keith Schengili-Roberts. I work for IXIASOFT as a DITA Specialist/Information Architect. And I like to write about DITA and the technical writing community. To get ahold of me you can email me at:

View all posts by