Graphics is a very important aspect of information presentation. Most formats, such as graphics interchange format (GIF), joint photographic experts group (JPEG), and portable network graphics (PNG), are raster based, but they do not describe the structure of the graphics. This paper is a research contribution to the description of graphical images recommended to computer science specialists.

It first analyzes scalable vector graphics (SVG) in depth, as a format for graphical description. This format allows many features, such as zooming, searching, and integrating with Web applications. Then, Fredj and Duce define a tool for graphical manipulation called graphical structure semantic markup languages (GraSSML). This is a layered approach: the semantic level is captured in extensible markup language (XML), and the ontology in Web ontology language (OWL). Another layer transforms the semantics into presentation level for human visualization. Each layer is described in detail. An example consists of an organizational chart with the textual content of each layer. An implementation of the system is presented with some dialogs captured.

The paper has an interesting subject with potential applicability. It is a medium-sized paper, but it gives enough information to stimulate the interest of the reader. Good references are included. The conclusion is quite short and does not suggest a lot of future work.