Object – Oriented Programming
Object – oriented programming is a relatively new approach to creating computer applications that seeks to address many of the problems with traditional programming techniques. The type of programming you have seen so far is known as functional (or procedural ) programming, often resulting in so – called monolithic applications, meaning all functionality is contained in a few modules of code (often just one). With OOP techniques, you often use many more modules of code, each offering specific functionality, and each module may be isolated or even completely independent of others. This modular method of programming gives you much more versatility and provides more opportunity for code reuse.
To illustrate this further, imagine that a high – performance application on your computer is a top – of – the – range racecar. Written with traditional programming techniques, this sports car is basically a single unit. If you want to improve this car, then you have to replace the whole unit by sending it back to the manufacturer and getting their expert mechanics to upgrade it, or by buying a new one. If OOP techniques are used, then you can simply buy a new engine from the manufacturer and follow their instructions to replace it yourself, rather than take a hacksaw to the bodywork.
In a more traditional application, the flow of execution is often simple and linear. Applications are loaded into memory, begin executing at point A, end at point B, and are then unloaded from memory. Along the way various other entities might be used, such as files on storage media, or the capabilities of a video card, but the main body of the processing occurs in one place. The code along the way is generally concerned with manipulating data through various mathematical and logical means. The methods of manipulation are usually quite simple, using basic types such as integers and Boolean values to build more complex representations of data.
With OOP, things are rarely so linear. Although the same results are achieved, the way of getting there is often very different. OOP techniques are firmly rooted in the structure and meaning of data, and the interaction between that data and other data. This usually means putting more effort into the design stages of a project, but it has the benefit of extensibility. After an agreement is made as to the representation of a specific type of data, that agreement can be worked into later versions of an application, and even entirely new applications. The fact that such an agreement exists can reduce development time dramatically. This explains how the racecar example works. The agreement here is how the code for the “ engine ” is structured, such that new code (for a new engine) can be substituted with ease, rather than requiring a trip back to the manufacturer. Conversely, it also means that the engine, once created, can be used for other purposes. You could put it in a different car, or use it to power a submarine, for example.
As well as providing an agreed – on approach to data representation, OOP programming often simplifies things by providing an agreement on the structure and usage of more abstract entities. For example, an agreement can be made not just on the format of data that should be used to send output to a device such as a printer, but also on the methods of data exchange with that device, including what instructions it understands, and so on. Going back to the racecar analogy, the agreement would include how the engine connects to the fuel tank, how it passes drive power to the wheels, and so on. As the name of the technology suggests, this is achieved using objects .
