Software Testing Methodologies Dichotomies

Importance of Modeling

What is Modeling?

·         A model is a simplification of reality

·         Modeling can be done for both simple and complex systems. It is more important for complex systems to understand them completely

·         A model may provide blueprints of a system that show both the structure and behavior

o   Structure means parts of the system and their relationships

For Example: A table has a top and legs. And, they are connected to form a table

o   Behavior defines the system's actions and how different parts contribute to it

Example: Learning is a part of college. Teachers teach and students learn. And the management organizes resources to make sure learning is carried out

Types of Models

• Different types of models are used in practice

For example

• o   A Curriculum Vitae (CV) is written in a textual model
• o   Representation of problems in the form of equations is a mathematical model
• o   Directions from one place to another are best shown as a map
• o   Parts of human body are best represented as a pictorial model
• o   A tabular model is a good way to capture details about a list of things
• o   Organization hierarchies are often represented as organization charts
• o   An algorithm is best understood in terms of a flow chart

Aims of modeling

A model:

·         Helps to visualize a system as it is or the way wanted

Example: Flowcharts help us to visualize an algorithm

·         Specifies the structure or behavior of a system

Example: Class diagrams in software systems show the system structure where as interaction diagrams show the behavior

A picture can be used to depict the structure of a human body

·         Gives a template that guides in constructing a system

Example: SRS for a software system

·         Document the decisions made

Example: An architecture document specifies the architectural decisions

Principles of Modeling

Principles of Modeling are:

Ø  Choice of Model is Important

Ø  Levels of Precision may differ

Ø  Connected to Reality

Ø  No Single Model is Sufficient

Choice of Model is Important:

Depending on the nature of the problem and the solution, we choose an appropriate type of model 

Levels of Precision May Differ:

Every model may be expressed at different levels of precision

The best kinds of models help to choose degree of detail. It depends on the person viewing the model and the need to view it

Best Models are connected to Reality:

All models simplify reality

Hence, we have to ensure that simplification does not hide any important details e.g. a simulation of a flight should consider different weather conditions

No Single Model is Sufficient:

Every complex system is approached by a set of independent models

We create models that can be built and studied separately, making sure that these models are still interrelated