Browse Prior Art Database

Visualization Compiler Disclosure Number: IPCOM000214369D
Publication Date: 2012-Jan-23
Document File: 3 page(s) / 29K

Publishing Venue

The Prior Art Database


Disclosed is a visualization compiler to help developers visualize an abstract model. The compiler incorporates four models: domain, graphics, render transform, and behavior. The approach makes notation definition and implementation of the corresponding fully functional diagrams much more cost effective, allowing visualization to be more applicable to high-volume, low-value use cases as opposed to limiting it to low-volume, high value cases.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Visualization Compiler

There are countless ways to visualize an abstract model. In diagramming software, developers typically use a notation model to describe a specific one; mapping model elements represent shapes, and introduce extra properties for the visual state such as position, size, and color. Defining notation models and implementing the corresponding functionality that bidirectionally binds them to abstract model instances as a functional diagram is expensive. Although partially driven by a desire for consistent interpretation, they are predominantly standardized and reused because of their excessive development cost.

Unfortunately, this approach stifles innovation and limits visualization to a handful of high-value use cases. Development of a single working diagram can engage several developers for months to years. Even while using the best frameworks available, the range of diagrams types is limited, and the frameworks are platform specific, so deploying to the web effectively means rewriting the framework as well.

The disclosed invention for a visualization compiler makes notation definition and implementation of the corresponding fully functional diagrams so cost effective it is nearly disposable, with the goal of shifting the economic balance away from monolithic reuse towards high-volume, low-value use cases. Coarsely, the solution repurposes a generic model transformation language for diagram notation definition, uses rule based inference to assemble interactive tools around it, and generates implementation code for a given target platform. The result is a visualization compiler that produces fully functional interactive diagrams with minimal user input, achieving an order of magnitude reduction in diagram development cost and complexity. The visualizations can be anything from sequence diagrams, to bar charts, to class diagrams -- all data-bound and interactive in a matter of hours by a single developer instead of months by several. While the current implementation deploys to the web, its model driven architecture should make extending it to deploy to other platforms theoretically cheaper than other frameworks.

To compare the invention for a visualization compiler to the state of the art Geographical Modeling Framework (GMF) diagram framework:

• GMF's notation model is static and centered on graph concepts (i.e., nodes and edges), so making diagrams that do not naturally fit into such a template. For example, sequence diagrams require an excessive amount of effort, often rendering particular diagram types out of reach in cost.

• GMF's notation implementation is fixed; it does not scale well to large visual...