Visual Language Representation

Visual Language Representation

What is a VPL?

•      A programming language that uses a visual representation (such as graphics, drawings, animation or icons, partially or completely)

•      A visual language manipulates visual information or supports visual interaction, or allows programming with visual expressions [Golin 90]

•      Any system where the user writes a program using two or more dimensions [Myers 90]

•      A visual language is a set of spatial arrangements of text-graphic symbols with a semantic interpretation that is used in carrying out communication actions in the world

What are NOT VPLs?

•      Visual Basic, Visual C++, Delphi, etc.

–     Still primarily textual languages with a graphical GUI builder

–     User interface portion of the language is visual, the rest is not

Motivations

•      Many people think in pictures.

•      Textual programming languages have proven to be difficult for many people to learn to use effectively.

•      Some applications are very well suited to graphical development approaches.

–     Scientific visualization

–     System simulation

Motivations

•      Many people think in pictures.

•      Textual programming languages have proven to be difficult for many people to learn to use effectively.

•      Some applications are very well suited to graphical development approaches.

–     Scientific visualization

–     System simulation

Example 1: Factorial

int Factorial(int n)

{

     if(n > 0) return (n*factorial(n-1));

     else

          return (1);

}

Classifications

1.     Purely visual languages

•         Icons or other graphical representations are manipulated

                    e.g. Cube, VIPR, Prograph, …

2.     Hybrid text and visual systems

•         Programs are created visually and then translated into an underlying textual language

•         Usage of graphical elements in an otherwise textual language

                   e.g. Rehearsal World

3.     Programming-by-example systems

•         Teach a system how to perform a task

                   e.g. Rehearsal World, Pygmalion

4.     Constraint-oriented systems

•         Popular for simulation design

                   e.g. ThingLab, ARK

5.     Form-based systems

•         Uses a spreadsheet metaphor

                   e.g. Forms/3, NoPumpG

What is Architectural Visualization?

•      Recall that we have characterized architecture as the set of principal design decisions made about a system

•      Recall also that models are artifacts that capture some or all of the design decisions that comprise an architecture

•      An architectural visualization defines how architectural models are depicted, and how stakeholders interact with those depictions

–     Two key aspects here:

•      Depiction is a picture or other visual representation of design decisions

•      Interaction mechanisms allow stakeholders to interact with design decisions in terms of the depiction

Models vs. Visualizations

•      It is easy to confuse models and visualizations because they are very closely related

•      In the previous lectures, we have not drawn out this distinction, but now we make it explicit

•      A model is just abstract information – a set of design decisions

•      Visualizations give those design decisions form: they let us depict those design decisions and interact with them in different ways

–     Because of the interaction aspect, visualizations are often active – they are both pictures AND tools

Different Relationships

Kinds of Visualizations: Textual Visualizations

•      Depict architectures through ordinary text files

–     Generally conform to some syntactic format, like programs conform to a language

–     May be natural language, in which case the format is defined by the spelling and grammar rules of the language

–     Decorative options

•      Fonts, colors, bold/italics

•      Tables, bulleted lists/outlines

Textual Visualizations: Interaction

•      Generally through an ordinary text editor or word processor

•      Some advanced mechanisms available

–     Syntax highlighting

–     Static checking

–     Autocomplete

–     Structural folding

Textual Visualizations

•      Advantages

–     Depict entire architecture in a single file

–     Good for linear or hierarchical structures

–     Hundreds of available editors

–     Substantial tool support if syntax is rigorous (e.g., defined in something like BNF)

•      Disadvantages

–     Can be overwhelming

–     Bad for graphlike organizations of information

–     Difficult to reorganize information meaningfully

–     Learning curve for syntax/semantics

Kinds of Visualizations: Graphical Visualizations

•      Depict architectures (primarily) as graphical symbols

–     Boxes, shapes, pictures, clip-art

–     Lines, arrows, other connectors

–     Photographic images

–     Regions, shading

–     2D or 3D

•      Generally conform to a symbolic syntax

–     But may also be ‘free-form’ and stylistic

Graphical Visualizations

Graphical Visualizations: Interaction

•      Generally graphical editors with point-and-click interfaces

–     Employ metaphors like scrolling, zooming, ‘drill-down’

•      Editors have varying levels of awareness for different target notations

–     For example, you can develop UML models in PowerPoint (or Photoshop), but the tools won’t help much

•      More exotic editors and interaction mechanisms exist in research

–     3D editors

–     “Sketching-based” editors

Graphical Visualizations

•      Advantages

–     Symbols, colors, and visual decorations more easily parsed by humans than structured text

–     Handle non-hierarchical relationships well

–     Diverse spatial interaction metaphors (scrolling, zooming) allow intuitive navigation

•      Disadvantages

–     Cost of building and maintaining tool support

•      Difficult to incorporate new semantics into existing tools

–     Do not scale as well as text to very large models

Hybrid Visualizations

•      Many visualizations are text-only

•      Few graphical notations are purely symbolic

–     Text labels, at a minimum

–     Annotations are generally textual as well

•      Some notations incorporate substantial parts that are mostly graphical alongside substantial parts that are mostly or wholly textual

Views, Viewpoints, & Visualizations

•      Recall that a view  is a subset of the  design decisions  in an architecture

•      And a viewpoint is the perspective from which a view is taken (i.e., the
filter that selects the subset)

•      Visualizations are associated with viewpoints

Advantages

•      Fewer programming concepts

•      Concreteness

•      Explicit depiction of relationships

•      Immediate visual feedback

•      Eliminates an intermediate step in the process of creating a program

Disadvantages

•      Deutsch Limit

•      The problem with visual programming is that you can't have more than 50 visual primitives on the screen at the same time.

•      Some situations in which text has superiority:

•      documentation,

•      naming to distinguish between elements that are of the same kind, and

•      expressing well-known and compact concepts that are inherently textual, e.g. algebraic formulas.