Course Content Overview

At a high level, this course involves four main components:

• Imperative problem solving
• C programming
• Controlling robots
• Storage and processing of data within a computer

In computing, we typically start with a problem that we wish to solve. Sometimes a problem is well-developed and nicely articulated, but often initial work is required to clarify just what needs to be done — in a real situation, what information does a user start with, and what results would be useful at the end.

Computer science highlights several distinct views of problem solving. Each perspective works extremely well for some types of problems, but not so well for other types of problems.

Imperative Problem Solving

This course focuses on imperative problem solving — an approach that is particularly useful in many areas, such as mathematics, statistics, science, engineering, computer systems, and computer networks.

Once a problem is clearly articulated, the goal of imperative problem solving is to develop a sequence of steps to solve the given problem. Typically, the problem begins with some information, just as in cooking, a chef starts with initial ingredients. From this starting point, we need to develop a sequence of step-by-step instructions for working with the data, just as a chef follows steps for cutting, blending, cooking ingredients, etc. In computer science, the resulting sequence of steps in called an algorithm, just as in cooking the steps are called a recipe.

When processing or cooking will be long and complex, imperative problem solving or cooking organizes instructions into high-level tasks, just as a recipe might indicate "mix all ingredients" or "heat while adding sugar until a firm consistency is obtained." Although the high-level tasks may suggest steps to be taken, some of these tasks may need to be subdivided into more detailed sub-tasks.

In some cases, both in imperative problem solving and in cooking, some tasks (e.g., printing numbers or separating egg whites from yolks) may be quite common, and we might develop one or more libraries that explain the details for handling those activities.

C Programming

When developing an algorithm or articulating the steps necessary to solve a problem, communication needs to be precise, clear, and unambiguous. If two people read the same description of an algorithm, they should have a common understanding of what to do and in what order. Further, when given starting data, reviewers should be able to apply the algorithm unambiguously to determine whether or not the result was correct. In addition, given two potential algorithms to solve a problem, the description of the algorithms should be sufficiently precise to allow analysis of whether one approach is more efficient or more accurate than another. Altogether, precision of language is essential for the specification of proposed solutions to problems.

Unfortunately, natural language, such as English, is neither precise nor unambiguous. Words may have multiple meanings, and general statements may not provide complete details. Since English or other natural languages are not adequate for technical communications of algorithms, new forms of exposition are essential in articulating algorithms for use in technical areas. Mathematical notation serves this purpose for some quantitative fields, but mathematical notation does not work particularly well for many problems for which we want help from a computer.

The C programming language has evolved over the years as a mechanism for precise communication for describing algorithms developed through imperative problem solving. The original version was developed between 1969 and 1973 by Dennis Ritchie with the intention of supporting research and development efforts at Bell Labs — then the research arm of American Telephone and Telegraph (AT&T). Since that time, C has been refined, expanded, and standardized. The first C Standard was approved in 1978 by the American National Standards Institute (ANSI). Subsequent standard versions have been approved by the International Organization for Standardization (ISO) in 1990, 1999, and 2011.

With international standards in place, programs written C that work on one machine can often be ported to another machine with little difficulty. Sometimes underlying hardware constraints may limit the movement of programs from one environment to another, but often C programs can be run without change on a wide variety of computers.

Although some extensions of Standard C may be limited to one machine environment or another, this course will emphasize features of Standard C. Programs in this course should work on any computer with access to the MyroC infrastructure and other widely-available libraries.

Controlling robots

Since this course emphasizes imperative problem solving and programming with the C programming language, the course must include a range of exercises that allow students to develop and sharpen their skills.

This course selects the control of robots as an integrating application theme. In particular, the course utilizes Scribbler 2 robots, because these devices come pre-packaged with the hardware needed for a wide range of applications. Workstations (running either Linux or Mac OS X) interact with the robots over Bluetooth, utilizing an infrastructure called MyroC. With the hardware and MyroC in place, the course can focus on imperative problem solving, which is a major theme of the course.

Although not all topics in the course relate directly to robots, many topics do. Overall, using C to control robots provides a rich area for practice and exploration using most of the concepts and techniques that need to be part of an introductory course on imperative problem solving.

Further, programming robots in C is fun!

Storage and processing of data within a computer

The C programming language provides a reasonably low-level view of the inner workings of a computer. In order to be effective, programmers must understand how data are organized within a computer and how various types of data are stored. Further, in some contexts, a programmer must explicitly allocate space for the storage of information and free that space when that information is no longer needed.

Whether working directly with C or not, an understanding of data storage and processing can provide insights into effective approaches for using the computer to solve problems. When one knows underlying principles of processing, one can organize subsequent work to take advantage of hardware, while avoiding approaches that may be less efficient or effective. One also may gain appreciation of what can go wrong in programs and how to respond.