Step 6. The last step is to add the data to the program and run the job completely. The computer will then perform the calculations 40
necessary to solve the problem. It will follow the instructions in the program to the minutest details. Therefore, one can say that the computer is a robot.It doesn't think, but simply does what it is told.
 Can a computer solve problems? Definitely not. It is a machine that 1
carries out the procedures which the programmer gives it. It is the programmer then who solves the problems. There are a few steps that one has to follow in problem solving:
 Step 1. The programmer must define the problem clearly. This means 5 that he or she has to determine, in a general way, how to solve the problem. Some problems are easy, while others take months of study. The programmer should always start by asking: 'Do I understand the problem?'.
 Step 2. The programmer must formulate an algorithm,which is a 10
straightforward sequence of steps of instructions used to solve the problem. Constructing an algorithm is the most important part of problem solving and is usually time-consuming. An algorithm can be described by a flowchart,which may be stated in terms of a sequence of precise sentences, or a block diagram.The latter is a diagrammatic 15 representation of the sequence of events to be followed in solving the problem. The relationship between the events is shown by means of a
connecting arrow------------- *. A block diagram can show if a process has
to be repeated or if there are alternative routes to be taken.
 Step 3. The programmer must translate the algorithm or flowchart 20
into a computer program. To do so, he or she writes detailed instructions for the computer, using one of the many computer languages available following the exact sequence of the flowchart algorithm. The program is usually written on coding sheetswhich have a specific format drawn on them. 25
Step 4. The programmer must then keypunch the program, or give the coding sheets to the keypunch operator to do it. The program is either punched on cards or, more usually, entered into the computer at a terminal with a visual display unit.
 Step 5. The program must then be tested. To do so, the computer 30
operator puts the deck of cards in the card reader and presses the 'read' button. This transfers the information to the memory of the computer. Alternatively, the program must be read from disk into the memory. Next, a printout shows if the program works or if it has errors (called bugs).If the programmer is using a terminal instead of 35 cards to enter the instructions it is possible, with the aid of a few commands, to store the program in the memory of the computer and get a printout.