Computer Science

Key Unit Competency
By the end of the unit, you should be able to explain programming paradigms.

Unit Outline
• Computer programming concepts.
• History of programming languages.
• High level programming languages.
• Computer programming paradigm.
• Features of good programming language.

Introduction
Computers have been applied in different areas, from controlling nuclear plants to providing games in mobile phones. Because of this diversity in computer use, a computer, tablet or mobile must have relevant programs. This unit introduces basic concepts used in computer programming, evolution of programming languages and programming paradigms since the advent of the first programmable machine.

Activity 9.1: Computer programming concepts
The structure of any language such as Kinyarwanda, Kiswahili, French, English or Chinese is described in terms of form (syntax) and meaning (semantic). In groups, research on the internet and use your knowledge in language studies to brainstorm on the two concepts.

9.1 Computer Programming Concepts
Before we begin discussing the details of computer programming, we need to consider a few concepts that will be used from time to time in the rest of this book. In this section, we briefly highlight some of the fundamental concepts used in programming which includes:

9.1.1 Computer program
A computer program refer to a set of instructions, written using a programming language to instruct a computer to perform a specified task. A program is like a recipe. It contains a list of ingredients (referred to as variables) and a list of instruction (statements) that tell the computer what to do with the variables.

9.1.2 Software
Though the term software and program are used interchangeably, technically, software refers to a program and associated documentations, while a program is basically a set of executable instructions loadable into computer memory.

9.1.3 Programming
Computer programming is a systematic process of writing a computer program using programming languages. The person who writes computer programs is referred to as a programmer. Other terms used to refer to a programmer are software developer and software engineer.

9.1.4 Programming languages
A programming language is a formal language that specifies syntax and semantics rules used in writing a computer program. Some examples of programming languages include BASIC, C, C++, Java, Pascal, FORTRAN and COBOL.

9.1.5 Source code
The term source code refers to a set of instructions or statements written by a programmer that are not yet translated into machine-readable form. A source code is mostly a text file written using programming languages like BASIC, Pascal, C or C++.

9.1.6 Object code
Once a source code is written, it can be translated into machine readable form referred to as object code. To translate source code statement to object code is similar to the way one can translate English to Kinyarwanda, there are language translators used to translate source code to object code.

9.1.7 Compilers and interpreters
A compiler is a language process that translates the entire source code into object code. The object file can be made into an executable program by carrying out another process known as linking. Linking combines compiled code with one or more existing object codes to create an execution file. In Windows operating system, you can easily identify an executable file because it has an EXE extension such as winword.exe. Unlike a compiler that translate the entire source code to object code, an interpreter translates source code one statement at a time. Because the interpreted statements are saved as an executable file, every time the program is run, each statement must be interpreted. Table 9.1 gives a summary of differences between compilers and interpreters.

Activity 9.2: Computer programming
Most students wonder how they would benefit from the study of mathematics and computer programming. In groups, brainstorm on some of the benefits of learning the two subjects and how the disciplines are interrelated.

Assessment Exercise 9.1
1. Define the terms:
(a) computer programming,
(b) source code.
(c) object code.
2. Differentiate between the compilers and interpreters.
3. Though the terms program and software are used interchangeably, they are technically different. Explain the difference between the two.

9.2 History of Programming languages
The person to be credited as the first programmer was a lady by the name Ada Byron in early 1800. Since then many programming languages have been developed over the years. These languages can be classified into two main categories and five generations. The first and second generations consist of low-level languages while the third to fifth generations comprise of high-level languages.

9.2.1 Low-level Programming Languages
Low-level languages are regarded as low because they can be directly understood by a computer while some requires minimal translation to machine readable form. Low level-languages are classified into two generations: first generation languages also known as machine languages, and second generation languages referred to as assembly languages.

9.2.2 First Generation Languages
First generation languages (1-GLs) refers to machine languages (binary code) used to program the first generation programmable computers such as UNIVAC and ENIAC. These computers were programmed by connecting wires on plug boards. The wiring configuration was used to represent data in binary form as a series of on’s (1) and off’s (0) in electronic circuits. Fig. 9.1 shows a sample binary code representing a program used to operate machines such as ENIAC.

NB: Machine programming was very slow, tedious and error prone. Furthermore, such a program is not portable because first electronic computers deferred from one another.

9.2.3 Second Generation Languages
The second generation languages (2-GLs) referred to as assembly languages marked the first successful attempt to make programming easier and faster. Most assembly languages allowed programmers to write programs as a set of symbolic codes known as mnemonics. Mnemonics are basically an abbreviation of keywords as shown in Fig. 9.2.

Unlike machine languages, program code written in assembly language has to be translated to machine code using a language processor known as assembler. An assembler is a special program that converts instructions written in low-level assembly code into machine code. Nevertheless, programs written using assembly languages are machine dependent hence not portable.

Activity 9.3: Second generation programming languages
In groups, research on the internet the programming languages used on Second generation computers such as IBM7094 and UNIVAC 1108.

9.2.4 Benefits and limitations of low-level languages
Having looked at the two categories of low-level programming languages, let’s highlight some of the benefits and limitations of low-level languages. Benefits
1. Program written using low level languages requires small amount of memory space.
2. The processor executes them faster because they require minimal or no translation.
3. Low level languages are stable and hardly crash or break down once written.

Limitations
1. Low level languages are difficult and cumbersome to use and learn.
2. They require highly trained experts both to develop and maintain.
3. Checking for errors (debugging) in low level programs is difficult and time consuming.
4. Low level programs are machine dependent hence they are not portable.

Assessment Exercise 9.2
1. Define the terms binary code, mnemonics, and assembler.
2. Differentiate between machine languages and assembly languages.
3. Explain how the first generation computers were programmed using binary code.
4. Highlight three advantages and three disadvantages of low level languages.
5. Mr. Kwizera bought a new electrical kettle. On the power switch of if were two inscribed digits 0 and 1:
(a) Explain what each of the two symbols stand for.
(b) Explain why the two symbols are important in computers and computer programming.

9.3 High-level Programming Languages
Due to drawbacks of low-level languages, high-level languages began to appear in 1950’s. High level languages that closely resembles natural (human) languages like English. Unlike low-level languages, high-level languages are independent of machine architecture. This means that, instead of a programmer spending more time learning the architecture of the underlying machine, more time is devoted towards solving a computing problem. Generally, high-level programming languages are classified into three generations namely: third generation (3-GLs), fourth generation (4-GLs), and fifth generation (5-GLs) programming languages.

9.3.1 Third generation languages
Third level languages (3-GLs) are also known as procedural or structured programming
languages. Procedural languages make it possible to break down a program into components known as procedures or modules each performing a particular task. Examples of 3-GL include Pascal, FORTRAN (Formula Translator), BASIC (Beginners All-Purpose Symbolic Instruction Code), C, C++, Adca and COBOL (Common Business Oriented Language).

9.3.2 Fourth generation languages
Fourth generation languages (4-GLs) were improvement on 3GLs meant to reduce
programming effort by making programming more easier and flexible.

Furthermore, most 4GLs incorporates advanced programming tools for integrating
programs with databases and generating summarised reports. Examples of 4-GLs
include Structured Query Language (SQL), Focus, PostScript, RPG II,
PowerBuilder, FoxPro, Python, Progress 4GL, and Visual Basic.

9.3.3 Fifth generation languages
Fifth generation languages (5-GLs) also known as natural languages are used to develop systems that solve problems using artificial intelligence. Artificial intelligence refers to computer systems that mimic human-like intelligence. Such intelligence include visual (seeing), perception, speech recognition, decision making and movement. Therefore, in 5GL programming, the programmer only worries about constraints required for the problem to be solved. Typical examples of 5GLs include Prolog, LISP, Scheme, Ocaml, and Mercury.

9.3.4 Benefits and limitations of high-level languages
Having looked at the various high-level programming languages, let’s highlight some of the benefits and limitations associated with most of these languages.

9.3.4.1 Benefits
1. High level languages are portable i.e. they are transferable from one computer
to another.
2. High level languages are user friendly and easy to use and learn.
3. High level languages are more flexible, hence they enhance the creativity of the
programmer and increase productivity in the workplace.
4. A program in high level languages is easier to correct errors.

9.3.4.2 Limitations
1. Their nature encourages use of many instructions in a word or statement hence the complexity of these instructions slows down program processing.
2. They have to be interpreted or compiled to machine readable form before the computer can execute them.
3. They require large computer memory to run.

Assessment Exercise 9.3
1. Distinguish between the following terms:
(a) Third generation.
(b) Fourth generation programming languages.
2. Briefly explain the evolution of programming languages. In each case, identify the generation and languages used.
3. State three advantages and three disadvantages of high-level languages.
4. On internet, conduct research to identify and discuss five examples of structured
programming language.

9.4 Computer Programming Paradigms
The term paradigm was first used by Thomas Kuhn in his 1962 to refer to theoretical
frameworks within which all scientific thinking and practices operate. In other words,
paradigm refers to theory or ideas concerning how something should be done, made,
or thought about. Paradigm shift refers to fundamental change on how something
should be done, made, or thought about.

9.4.1 Definition of Programming Paradigm
Programming paradigm refers to pattern, theory or systems of ideas that are used to
guide development of computer programs. In other words, it is a school of thought or
philosophy that defines concepts, practices and views on how computer programming
should be conceptualized or performed. Several programming paradigms have
evolved each of which presents programmers with a specific mode of thinking about
computer programming. In the next section, we classify programming paradigms into
imperative, functional, logic and object oriented.

9.4.2 Classification of Programming Paradigms
Programming paradigm may be classified into four main categories namely imperative programming, functional programming, logic programming and object-oriented programming.

9.4.2.1 Imperative programming paradigm
Imperative programming also referred to as procedure-oriented is a paradigm in which commands (program instructions) are executed in sequential order. One of the fundamental characteristic of programs written using imperative languages is that they have variables that change during program execution. For example, consider the following statement that adds two numbers x and y and assigns the result to a variable named sum: sum = x + y
Every time different values for variables x and y are provided, sum changes from the previous state to new state as shown in Table 9.2.

Programming languages that support imperative programming including machine languages, assembly languages, Basic, Pascal and C.

9.4.2.2 Functional Programming Paradigm
Functional programming is a paradigm based on concept of functions that consists of the function name and list of values known as parameters enclosed in parenthesis. The main difference between functional programming and imperative paradigm is that functional programming does not require use of assignment statements to manipulate variables. Instead, manipulation of variables is accomplished by applying functions to a list of parameters also known as arguments. The following syntax known as polish notation is used to represent a function and list of arguments:

(function_name parameter1... parametern);

For example, consider a function that calculates sum of four parameters 5, 4, 7 and
9. We can use addition symbol (+) or mnemonic add to represent addition function as follows:
(+ 5 4 7 9) or (Add 5 4 7 9)
In this case, the function takes four parameters to calculate the total; this gives us 25. The parameters in this example can also be manipulated using other arithmetic functions like subtraction (-), multiplication (*) and division (/). Examples of programming languages that support functional paradigm include LISP, Scheme, Haskell, MetaLanguage (ML), Miranda, Caml, and F#.

Activity 9.4: Programming paradigms
Using examples, differentiate between imperative, and functional programming paradigms.
In groups, brainstorm on benefits and limitations of functional programming paradigm. Use a sample function in Polish notation to demonstrate how to subtract or multiply three parameters.

9.4.2.3 Logic Programming Paradigm
Logic programming is a rule-based paradigm that focuses on use of logic or predicate calculus. In logic programming paradigms, only facts and rules are declared to produce desired results. This means that a logic program is a set of facts that make use of a set of rules to answer a query. For example, the following statement in a language known as Prolog (standards for programming logic) could mean that if ann is the mother of shella, then ann is an ancestor of shella:

ancestor(ann, shella) :- mother(ann, shella).

Logic programming paradigm fits well when applied in artificial intelligence (AI) that deal with the extraction of knowledge from basic facts and rules. In artificial intelligence, various logical assertions (proportions) about a situation are made to establish all known facts. Languages that emphasize logic programming paradigm include Prolog, GHC, Parlog, Vulcan, Polka and Mercury.

Activity 9.5: Logic programming
• In groups, brainstorm on benefits and limitations of logic programming paradigm.
• Use a sample functional program to demonstrate how rule-based program
statements are executed in regard to facts, rules, inference and answers to queries.

9.4.2.4 Object Oriented Programming Paradigm
Object-Oriented Programming Paradigm (OOP) is the latest paradigm in which properties (data) and operations (procedures) are combined to form objects. Therefore, an object represents a real-world “thing” such as a person, animal, plant, place, or building. In object-oriented programming, similar objects are grouped together to form classes. For example, the Table 9.3 below shows three types of classes that define properties and operations applicable to each object:

Because the latest paradigm shift is development of OPP programs, most imperative languages like C, Pascal and Basic have evolved to support OOP. Examples of programming languages that support OOP include Delphi Pascal, C++, Java, C#, Visual Basic.Net, and Objective-C.
In summary, Table 9.4 shows the four major programming paradigms namely
imperative, functional, logic, and object-oriented programming:

Activity 9.6: OOP Paradigm
• Though OOP differs significantly from procedure-oriented paradigm, we may not consider this category separate from imperative paradigms. This is because some object-oriented languages such as F# supports functional programming. Using valid references, defend or challenge this argument.
• In groups, discuss the terms classes, inheritance and polymorphism. In addition, brainstorm on benefits and limitations of object-oriented programming over imperative programming paradigm.

9.5 Features of Good Programming Language
Criteria for evaluating programming languages and paradigms may be controversial but Sebesta in his book, “Concepts of Programming Languages, tenth edition” suggests four main criteria namely: readability, writability, reliability and cost.
• Overall simplicity: Overall simplicity of a programming language influences its ease of learning and readability.
• Good orthogonality: Relatively small set of simple constructs can be combined in a number of ways to provide required control and data structures of the language. Limited orthogonalilty makes it easier to learn, read, and understand a language.
• Adequate data types and data structures: Presence of adequate facilities for defining data types and data structures help increase the readability of a programming language.
• Clear syntax design: The syntax, or form, of the elements of a language has a significant effect on the readability of programs. For example, use of special words such as end if makes a program more readable.
• Support for abstraction: Programming language should provide facilities to define and then use complicated structures or operations in ways that allow many of the details to be ignored. Two types of abstraction are process (subprograms) and data abstraction (structures, records, objects).

• Expressivity: Typically expressivity means that a language has convenient ways of
specifying computations. For example, in C, C++ and Java, the notation count++ is a more convenient and shorter way of incrementing count by 1 equivalent to count = count + 1.
• Mechanisms to handle exceptions: This is the ability of a program to intercept run-time errors or detect other unusual conditions, take corrective measures, and then continue with normal execution. A good programming language should provide mechanism to handle exceptions.
• Type checking: Type checking refers to testing for data type errors during program compilation or run-time (execution). Because run-time type checking is expensive, it is more desirable for a programming language to verify data type at compilt-time.
• Cost-effective: The total cost of a programming language can be evaluated in terms
of compiler cost, software development process, compilation time, implementation
platforms, programmer training and maintenance.

Activity 9.7: Qualities of a good program
Research on internet and write brief report on characteristics of a good programming language.

Exercise 9.4
1. Explain the concepts: object-orientation, and logic programming paradigms.
2. Explain why knowledge of programming language characteristics can benefit thewhole computing community.
3. Explain the programming paradigm supported by F# programming language.
4. Explain why is it useful for a programmer to have some background in language
design, even though he or she may never actually design a programming language?

Unit Test 9
1. Differentiate between a computer program and software.
2. Explain how evolution of computers have influenced paradigm shift in computer programming.
3. List three examples of object-oriented programming languages.
4. Differentiate between procedural programming and functional programming
paradigms.
5. Pascal and FORTRAN are examples of _______ generation programming
languages.
6. Procedural languages make it possible to break down a program into components
known as ___________ or ________.
7. A programming paradigm in which a program is executed in sequenced order
is known as _________.