For every 25% increase in problem complexity, there is a 100% increase in complexity of the software solution.

As new software engineering techniques help us to build larger, more complex systems, the demands change. Systems have to be built and delivered more quickly. Larger, even more complex systems are required. Systems have to have new capabilities that were previously thought to be impossible.

The aim of the design workflow is to refine the artifacts of the analysis workflow until the material is in a form that can be implemented by the programmers.

a complex, iterative process. The initial design solution will likely be wrong and certainly not optimal.

The aim of the design workflow is to refine the artifacts of the analysis workflow until the material is in a form that can be implemented by the programmers.

The process of resolving a functional relationship into its constituent parts

There is seldom one best design solution to a software problem.

focus on nonfunctional requirements • decomposition into modules - module = function? class? package?

a complex, iterative process. The initial design solution will likely be wrong and certainly not optimal.

Code-and-fix Functional decomposition Structured programming Object-oriented programming, design, and requirements analysis Design patterns and tactics

There is seldom one best design solution to a software problem.

a complex, iterative process. The initial design solution will likely be wrong and certainly not optimal.

Top designers pursue a design solution not by working in an ordered and top-down (or even bottom-up) fashion, but by pursuing targets of important opportunity. And those targets of important opportunity are usually the difficult ones, the ones for which the designer sees no immediate solution. When expert designers got right down to the nitty-gritty of design work, their approaches turned out to be heuristic, trial and error.

There is seldom one best design solution to a software problem.

Code-and-fix Functional decomposition Structured programming Object-oriented programming, design, and requirements analysis Design patterns and tactics

The process of resolving a functional relationship into its constituent parts

Code-and-fix Functional decomposition Structured programming Object-oriented programming, design, and requirements analysis Design patterns and tactics

There is seldom one best design solution to a software problem. In a room of top software designers, if any two of them agree, that’s a majority. Bill Curtis. The design process is opportunistic.

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier (Yourdon 1979)

The process of resolving a functional relationship into its constituent parts

Code-and-fix Functional decomposition Structured programming Object-oriented programming, design, and requirements analysis Design patterns and tactics

A module is any collection of executable program statements meeting all of the following criteria

The process of resolving a functional relationship into its constituent parts

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier (Yourdon 1979)

reductionism (vs holism) - decomposition, reductionism, stepwise refinement, divide and conquer, top-down design, etc. • hierarchy • modularity • information hiding - hide information about the design from other modules, at the points of difficulty or likely change - abstraction, encapsulation

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier

A module is any collection of executable program statements meeting all of the following criteria:

On the Criteria to be Used in Decomposing Systems into Modules,” Communications of the ACM, December 1972.

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier

A module is any collection of executable program statements meeting all of the following criteria

A module is any collection of executable program statements meeting all of the following criteria

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier

avoid tests and gotos that lead to "spaghetti code"

avoid tests and gotos that lead to "spaghetti code"

A lexically contiguous sequence of program statements, bounded by boundary elements, with an aggregate identifier

A module is any collection of executable program statements meeting all of the following criteria

it is a closed subroutine • it can be called from any other module in the program • it has the potential of being independently compiled

avoid tests and gotos that lead to "spaghetti code

A programming paradigm aimed at improving the clarity, quality, and development time of a computer program by making extensive use of subroutines, block structures, for and while loops

The art of designing the components of a system and the interrelationships between those components in the best possible way

it was the use of the three basic control structures, single-entry/single-exit modules, and the use of indented code.

the use of the three basic control structures, short routine lengths, and indented coding.

the three basic control structures, application of management techniques to top-down design and implementation, and levels of abstraction.

"a top-down sequence for program unit creation and testing, and a technical standard for the coding of each unit

the use of the three basic control structures, single-entry/single-exit modules, and the use of indented code.

"a top-down sequence for program unit creation and testing, and a technical standard for the coding of each unit

as the use of the three basic control structures, short routine lengths, and indented coding.

it as the three basic control structures, application of management techniques to top-down design and implementation, and levels of abstraction.

"a top-down sequence for program unit creation and testing, and a technical standard for the coding of each unit

the three basic control structures, application of management techniques to top-down design and implementation, and levels of abstraction.

as the use of the three basic control structures, short routine lengths, and indented coding.

s the use of the three basic control structures, single-entry/single-exit modules, and the use of indented code.

"a top-down sequence for program unit creation and testing, and a technical standard for the coding of each unit

the three basic control structures, application of management techniques to top-down design and implementation, and levels of abstraction.

the use of the three basic control structures, short routine lengths, and indented coding.

the use of the three basic control structures, single-entry/single-exit modules, and the use of indented code.

The structure of a product must conform to the structure of its data

The art of designing the components of a system and the interrelationships between those components in the best possible way The process of deciding which components interconnected in which way will solve some well-specified problem.

Any algorithm can be expressed using only three control structures. • executing one subprogram, and then another subprogram (sequence) • executing one of two subprograms according to the value of a Boolean expression (selection) • executing a subprogram until a Boolean expression is true (iteration)

the use of the three basic control structures, single-entry/single-exit modules, and the use of indented code.

The art of designing the components of a system and the interrelationships between those components in the best possible way. The process of deciding which components interconnected in which way will solve some well-specified problem.

Using three control structures (sequential, iteration, selection) • Single entry / single exit - Does not use GOTOs • Indented code • Short functions (fit on 1-2 pages) • High cohesion, low coupling, …

The structure of a product must conform to the structure of its data

Any algorithm can be expressed using only three control structures. • executing one subprogram, and then another subprogram (sequence) • executing one of two subprograms according to the value of a Boolean expression (selection) • executing a subprogram until a Boolean expression is true (iteration)

has never been as popular as operationoriented design • with the rise of OOD, data-oriented design has largely fallen out of fashion

The structure of a product must conform to the structure of its data

• has never been as popular as operationoriented design • with the rise of OOD, data-oriented design has largely fallen out of fashion

Michael Jackson [1975] • Jean-Dominique Warnier [1976] • Kenneth Orr [1981]

The structure of a product must conform to the structure of its data

• has never been as popular as operationoriented design • with the rise of OOD, data-oriented design has largely fallen out of fashion

The structure of a product must conform to the structure of its data

• has never been as popular as operationoriented design • with the rise of OOD, data-oriented design has largely fallen out of fashion

Michael Jackson [1975] • Jean-Dominique Warnier [1976] • Kenneth Orr [1981]

aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

Represents concepts as “objects” that have • data (attributes that describe the object) • associated procedures known as methods Objects are usually instances of classes. Data is accessed by calling specially written functions, commonly called methods. - Objects encapsulate their data within a set of functions designed to ensure that the data are used appropriately.

Are packages of design decisions that have been found to work repeatedly in practice Have known properties that permit reuse Describe classes of design solutions • “chunk” design solutions into a single concept Patterns are found in practice by definition • one does not invent patterns – one discovers them • patterns spontaneously emerge in reaction to environmental conditions - as long as conditions change, new patterns will emerge

Are packages of design decisions that have been found to work repeatedly in practice Have known properties that permit reuse Describe classes of design solutions • “chunk” design solutions into a single concept Patterns are found in practice by definition • one does not invent patterns – one discovers them • patterns spontaneously emerge in reaction to environmental conditions - as long as conditions change, new patterns will emerge

Information hiding, low coupling, high cohesion are the foundation for the object-oriented paradigm. Represents concepts as “objects” that have • data (attributes that describe the object) • associated procedures known as methods Objects are usually instances of classes. Data is accessed by calling specially written functions, commonly called methods. - Objects encapsulate their data within a set of functions designed to ensure that the data are used appropriately.

- aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

solving ever more complex problems  building systems that can be maintained

hide details that may change

solving ever more complex problems  building systems that can be maintained

aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

hide details that may change

solving ever more complex problems  building systems that can be maintained

aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

hide details that may change

chunking into higher-level concepts

Focused on one thing

minimize interconnections

chunking into higher-level concepts

chunking into higher-level concepts

focused on one thing

minimize interconnections

Partitioning the system into parts having identifiable and understandable boundaries

Representing the system as a hierarchy

Maximizing the independence among the parts of the system

aka stepwise refinement, reductionism (vs wholism) - Comprehensible (Understandable) (Module size): small enough to understand (Miller’s Law 7 ± 2) - Separation of Concerns

hide details that may change

Divide-and-conquer - Boundaries  Interfaces - Boundaries  Abstraction, Encapsulation

Allows us to iteratively separate our ideas and to deal with increasing amounts of detail - Design tree

Module independence - Low coupling

Divide-and-conquer - Boundaries  Interfaces - Boundaries  Abstraction, Encapsulation

Allows us to iteratively separate our ideas and to deal with increasing amounts of detail - Design tree

Module independence - Low coupling

Divide-and-conquer - Boundaries  Interfaces - Boundaries  Abstraction, Encapsulation

Allows us to iteratively separate our ideas and to deal with increasing amounts of detail - Design tree

Module independence - Low coupling

The most important consideration in good design, and the single idea on which most of composite design is based, is the idea of high module independence. - To have achieved high independence, the system should have been partitioned so that if two components were not closely related, they would fall into separate partitions, and if two components were closely related, they would fall into the same partition. The key to successful program structuring, therefore, is maximizing module independence.

recursively breaking down a problem into two or more sub-problems of the same or related type, until these become simple enough to be solved directly - aka stepwise refinement, decomposition, functional decomposition, top-down design, reductionism

decompose until “simple enough”? - divided into modules small enough for the programmer to easily understand - one or two pages (no more than 100 lines of codes) is a reasonable rule of thumb - very small modules (less than 10 lines of code) are recommended by many experts

recursively breaking down a problem into two or more sub-problems of the same or related type, until these become simple enough to be solved directly - aka stepwise refinement, decomposition, functional decomposition, top-down design, reductionism

- decompose until “simple enough”? - divided into modules small enough for the programmer to easily understand - one or two pages (no more than 100 lines of codes) is a reasonable rule of thumb - very small modules (less than 10 lines of code) are recommended by many experts

If the subtasks are not truly independent of one another, then we are not solving just the two subtasks – for in the solution of non-independent parts, we also are simultaneously dealing with some aspects of the other parts.

recursively breaking down a problem into two or more sub-problems of the same or related type, until these become simple enough to be solved directly - aka stepwise refinement, decomposition, functional decomposition, top-down design, reductionism

decompose until “simple enough”? - divided into modules small enough for the programmer to easily understand - one or two pages (no more than 100 lines of codes) is a reasonable rule of thumb - very small modules (less than 10 lines of code) are recommended by many experts

If the subtasks are not truly independent of one another, then we are not solving just the two subtasks – for in the solution of non-independent parts, we also are simultaneously dealing with some aspects of the other parts.

A hierarchical approach is an aid because it allows us to iteratively separate our ideas and to deal with increasing amounts of detail. Hierarchical structure • partial ordering (uses or depends on) between modules or programs • “design tree”

A program that embodies Separation of Concerns well is called a modular program. • Modularity, and hence separation of concerns, is achieved by encapsulating information inside a section of code that has a well-defined interface. • Encapsulation is a means of information hiding. - Wikipedia “Separation of concerns” • Many other design principles derive from SoC

“Clean” decomposition • every module is characterized by its knowledge of a design decision which it hides from all others • “information hiding” and its derivatives

A hierarchical approach is an aid because it allows us to iteratively separate our ideas and to deal with increasing amounts of detail. Hierarchical structure • partial ordering (uses or depends on) between modules or programs • “design tree”

A program that embodies Separation of Concerns well is called a modular program. • Modularity, and hence separation of concerns, is achieved by encapsulating information inside a section of code that has a well-defined interface. • Encapsulation is a means of information hiding. - Wikipedia “Separation of concerns” • Many other design principles derive from SoC

“Clean” decomposition • every module is characterized by its knowledge of a design decision which it hides from all others • “information hiding” and its derivatives

A hierarchical approach is an aid because it allows us to iteratively separate our ideas and to deal with increasing amounts of detail. Hierarchical structure • partial ordering (uses or depends on) between modules or programs • “design tree”

A program that embodies Separation of Concerns well is called a modular program. • Modularity, and hence separation of concerns, is achieved by encapsulating information inside a section of code that has a well-defined interface. • Encapsulation is a means of information hiding. - Wikipedia “Separation of concerns” • Many other design principles derive from SoC

“Clean” decomposition • every module is characterized by its knowledge of a design decision which it hides from all others • “information hiding” and its derivatives

The principle of segregation of the design decisions in a computer program that are most likely to change, thus protecting other parts of the program from extensive modification if the design decision is changed • provide a stable interface which protects the remainder of the program from the implementation (the details that are most likely to change) • the ability to prevent certain aspects of a class or software component from being accessible to its clients

Generalization by reducing the information content of a concept or an observable phenomenon, typically in order to retain only information which is relevant for a particular purpose

define a procedure — extend the language • abstraction to functions (procedures, methods, routines, subroutines)

designer thinks at the level of an abstract data type • abstraction to classes (abstract data types)

Generalization by reducing the information content of a concept or an observable phenomenon, typically in order to retain only information which is relevant for a particular purpose

define a procedure — extend the language • abstraction to functions (procedures, methods, routines, subroutines)

designer thinks at the level of an abstract data type • abstraction to classes (abstract data types)

define a procedure — extend the language • abstraction to functions (procedures, methods, routines, subroutines)

designer thinks at the level of an abstract data type • abstraction to classes (abstract data types)

Used to hide the values or state of a structured data object inside a class A language mechanism for restricting direct access to some of the object's components

localizes data and prevents them from being directly exposed to the rest of the system • enhances quality and reuse because the data are only accessible through calls to the operations (methods or functions) of a class

the degree to which the elements of a module belong together - focusing on a single concept - cohesion measures the strength of relationship between pieces of functionality within a given module

the manner and degree of interdependence between software modules - a measure of how closely connected two routines or modules are - the strength of the relationships between modules

the degree to which the elements of a module belong together - focusing on a single concept - cohesion measures the strength of relationship between pieces of functionality within a given module

the manner and degree of interdependence between software modules - a measure of how closely connected two routines or modules are - the strength of the relationships between modules

The degree of interaction between two modules The degree of interdependence between software modules A measure of how closely connected two routines or modules are The strength of the relationships between modules

The degree of interaction between two modules The degree of interdependence between software modules A measure of how closely connected two routines or modules are The strength of the relationships between modules

5. Data coupling

4. Stamp coupling

3. Control coupling

2. Common coupling

1. Content coupling

Two modules are data coupled if all parameters are homogeneous data items (simple parameters, or data structures all of whose elements are used by called module). - when modules share data through, for example, parameters - each datum is an elementary piece, and these are the only data shared (e.g., passing an integer to a function that computes a square root) (Wikipedia)

Data coupling is the only good kind of coupling. • maximizes information hiding • follows the principle of least privilege The other kinds of coupling compromise information hiding to one degree or another.

A change in one module usually forces a ripple effect of changes in other modules. Assembly of modules may require more effort and/or time due to the increased inter-module dependency. A particular module may be harder to reuse and/or test because dependent modules must be included.

the number of (non-unique) modules invoked by this module - high fan-out (more than about seven) suggests an overly complex class  bad (McConnell 2004) - Green zone -- Up to fan out 6 (Tockey 2018) - Yellow zone -- Fan out 7-9 - Red zone -- Fan out 10 or greater

Data coupling is the only good kind of coupling. • maximizes information hiding • follows the principle of least privilege The other kinds of coupling compromise information hiding to one degree or another.

A change in one module usually forces a ripple effect of changes in other modules. Assembly of modules may require more effort and/or time due to the increased inter-module dependency. A particular module may be harder to reuse and/or test because dependent modules must be included.

the number of (non-unique) modules invoked by this module - high fan-out (more than about seven) suggests an overly complex class  bad (McConnell 2004) - Green zone -- Up to fan out 6 (Tockey 2018) - Yellow zone -- Fan out 7-9 - Red zone -- Fan out 10 or greater

The number of modules invoked by this module

the number of occurrences of other modules invoking this module

number of other classes that is a class is coupled to

Up to fan out 6(Tockey 2018)

Fan out 7-9

Fan out 10 or greater

Up to fan out 6(Tockey 2018)

Fan out 7-9

Fan out 10 or greater

Up to fan out 6(Tockey 2018)

Fan out 7-9

Fan out 10 or greater

The number of (non-unique) modules invoked by this module

the number of occurrences of other modules invoking this module

number of other classes that a class is coupled to

number of other classes that a class is coupled to

number of classes that directly inherit one class

Maximum number of nodes between root and lowest node in the hierarchy

number of methods a class is accessible to, including methods implemented in own class as well as methods accessible due to inheritance

number of other classes that directly inherit one class

number of other classes that a class is coupled to

Maximum number of nodes between root and lowest node in the hierarchy

maximum number of nodes between root and lowest node in the hierarchy

number of other classes that a class is coupled to

number of classes that directly inherit one class

number of methods a class is accessible to, including methods implemented in own class as well as methods accessible due to inheritance

number of classes that directly inherit one class

maximum number of nodes between root and lowest node in the hierarchy

a measure of the strength of relationship between the methods and data of a class and some unifying purpose or concept served by that class (Wikipedia)

• focusing on a single thing • having a single responsibility - knowing something - doing something

the probability that a change scenario that affects a responsibility will also affect other (different) responsibilities - the higher the cohesion, the lower the probability that a given change will affect multiple responsibilities

the probability that a change scenario that affects a responsibility will also affect other (different) responsibilities - the higher the cohesion, the lower the probability that a given change will affect multiple responsibilities

• focusing on a single thing • having a single responsibility - knowing something - doing something

a measure of the strength of relationship between the methods and data of a class and some unifying purpose or concept served by that class (Wikipedia)

a measure of the strength of relationship between the methods and data of a class and some unifying purpose or concept served by that class (Wikipedia)

focusing on a single thing • having a single responsibility - knowing something - doing something

- the probability that a change scenario that affects a responsibility will also affect other (different) responsibilities - the higher the cohesion, the lower the probability that a given change will affect multiple responsibilities

when parts of a module are grouped because they are logically categorized to do the same thing even though they are different by nature (e.g. grouping all mouse and keyboard input handling routines) (Wikipedia)

it performs multiple, completely unrelated actions.

when parts of a module are grouped by when they are processed; the parts are processed at a particular time in program execution (e.g. a function which is called after catching an exception which closes open files, creates an error log, and notifies the user) (Wikipedia)

it performs a series of related actions, one of which is selected by the calling module.

it performs multiple, completely unrelated actions.

it performs a series of actions related in time.

it performs a series of related actions, one of which is selected by the calling module.

it performs multiple, completely unrelated actions.

it performs a series of actions related in time.

it performs a series of actions related in time.

it performs a series of actions related by the procedure to be followed by the product

it performs a series of related actions, one of which is selected by the calling module.

it performs a series of actions related by the procedure to be followed by the product, but in addition all the actions operate on the same data.

it performs a series of actions related by the procedure to be followed by the product, but in addition all the actions operate on the same data.

it performs a series of actions related in time.

It performs a number of actions, each with its own entry point, with independent code for each action, all performed on the same data structure.

if it performs a series of actions related by the procedure to be followed by the product, but in addition all the actions operate on the same data.

it performs a series of actions related by the procedure to be followed by the product.

Performs exactly one action

More reusable

Corrective maintenance is easier

Easier to extend a product

When parts of a module are grouped because the output from one part is the input to another part like an assembly

it performs a number of actions, each with its own entry point, with independent code for each action, all performed on the same data structure

When parts of a module are grouped because the output from one part is the input to another part like an assembly line

More reusable

Corrective maintenance is easier

Easier to extend a product

logical cohesion

communicational & sequential cohesion are good • functional & informational cohesion are superior

which parts of a module are grouped because the output from one part is the input to another part like an assembly line

communicational & sequential cohesion are good • functional & informational cohesion are superior

logical cohesion

temporal cohesion

communication cohesion

Temporal cohesion

communicational cohesion

logical cohesion

logical cohesion

you get modules like initialize, Process, and Terminate

Temporal cohesion

a start toward defining abstract data types

communicational cohesion

Sequential Cohesion

a start toward defining abstract data types

Communicational cohesion

sequential cohesion

logical cohesion

You get modules like initialize, process, and terminate

TeTemporal cohesion

logical cohesion

Sequential cohesion

a start toward defining abstract data types

communicational cohesion

simpler, having fewer operations

Logical changes in the domain affect fewer modules

Changes in one module require fewer changes in other modules

application developers will find the component they need more easily among the cohesive set of operations provided by the module

simpler, having fewer operations

logical changes in the domain affect fewer modules

Changes in one module require fewer changes in other modules

application developers will find the component they need more easily among the cohesive set of operations provided by the module

application developers will find the component they need more easily among the cohesive set of operations provided by the module

logical changes in the domain affect fewer modules

changes in one module require fewer changes in other modules

application developers will find the component they need more easily among the cohesive set of operations provided by the module

Too many parameters to a function suggests that the function is trying to do too much

Keep unit interfaces small by limiting the number of parameters on the function to no more than 4

up to 4 parameters

5 or 6 parameters

7 or more

5 or 6 parameters

up to 4 parameters

7 or more

5 or 6 parameters

7 or more

up to 4 parameters

up to 4 parameters

5 or 6 parameters

7 or more

The ability of an instantiated object (at runtime) to understand and interpret the message sent from a calling object.

is the starting point of object oriented paradigm

results from classes being generalized into higher-level or abstracted classes.

results from classes being generalized into higher-level or abstracted classes

The ability of an instantiated object to understand and interpret the message sent from a calling object

Is the starting point of object oriented paradigm based on the requirements appearing in the problem space

fixing defects

changing environment-porting to new platforms

improve performance or maintainability, including enhancements

detect and correct latent faults in the software product refactoring!