Abstract: |
The main purpose of the work was to show how, under certain conditions and in some application domains, visual programming languages can be usefully exploited to improve the interaction of advanced users with the computer, achieving, with little effort, even complex functionalities. Although the debate about the usefulness of visual programming languages compared with textual ones is quite far from being over, it is indisputable that, at least for certain applications, interacting with objects placed in a two-dimensional space may be extremely worthwhile. Graphic elements, in fact, have the advantage of being characterized by shape, dimension, position and possibly color, all attributes which may help better understand the meaning of what is displayed on the screen. As far as we are concerned, we believe that visual programming, if properly exploited and provided with adequate control structures, holds very great potential and can speed up reasoning processes.
Visual programming, in general, means using visual representations to communicate data and operations to the computer. In particular, we focused our attention on data-flow visual programming languages, characterized by a simple yet powerful computational paradigm. However, given that the lack of adequate iterative structures has often proved to be one main shortcoming of these languages, another important purpose of the thesis was to propose some new solutions for implementing iterations, analyzed in detail, with the VIPERS visual language.
The application point of view has been tackled with special reference to the Internet and the World Wide Web, where one is most likely to find advanced users with the need to reach, in simple and fast manners, objectives that are hardly obtainable through the ordinary search and navigation tools. Remaining in the application sphere, a data-flow visual approach to symbolic computing has also been proposed.
Outside the data-flow programming domain but still within the scope of visual languages, integration of the concepts of visual language and graphic interface may create synergies thanks to which planning the execution of a certain set of operations becomes more immediate. In this context, a visual system to specify mission plans for a submarine robot has been proposed.
Finally, since any “tool” is useful only if it is sufficiently usable, the problem of carrying out usability tests for visual structures has been tackled, with particular reference to the data-flow model and iterative control constructs. A simple but effective Web-based test methodology has also been proposed, which can be applied whenever the interaction between the user and complex images is being studied. |
