This topical outline and syllabus is a work in progress. We have been thinking about these issues for a while and decided it was time to turn our intuitions into something more rigorous and practical. We started out with a simplistic view of how applications might be based on models, and expected as a result to be able to organize this course around different categories of model-based applications (e-books, e-forms, etc.). In our work to develop this course we've been amazed by the complexity of the problem, overwhelmed by the amount of research that has been conducted, but encouraged by the gap between research and commercial efforts -- because that suggests that we can learn things and exploit what we learn, even in a short semester course. Nevertheless, it all means that we are probably going to focus on the varieties of research perspectives and not have as clear an organization on the application end.

So our revised course goals are:

• Acquire concepts and vocabulary for thinking about model-based user interfaces.
• Get some overall familiarity with the different research perspectives and study a few efforts in more detail.
• Get some overall familiarity with different commercial/tool approaches and study a few efforts in more detail.
• Get some overall familiarity with XML vocabularies for specifying user interface models
• Analyze the differences between "inside out" or "model-based" approaches to user interface development and those that are more "outside in" in which models are at best implicit.
• Work toward a hybrid or harmonized approach that shrinks or eliminates the gap between these two approaches by harvesting the most promising "nuggets" that we can embed in Document Engineering and User Interface design as taught at SIMS.
• Use XForms, OXF, and the "Rules-Based Infrastructure" work of Charles and Daly as a conceptual and technical testbed.