WHAT IS
SOFTWARE?

Baking cakes has so much in common with software that we hardly need anything further to explain the central ideas behind software. Cake recipes describe which ingredients are processed in which order and how, up to the point when the cake is ready: Mix eggs, sugar and butter, add flour and baking powder, knead the dough and then bake it at 160 degrees for 30 minutes. There are the ingredients, the individual steps of preparation and eventually a cake. Also needed are tools that more than likely will not be mentioned in the recipe: Mixer, mixing bowl and oven. There are also the baker and the author of the recipe. Then there are the steps in the recipe, namely mixing, stirring, waiting and baking. In this case, you are in the role of bakers, carrying out these individual steps in the right order, ensuring that the next step can be taken using results from the previous step. With that, we now already have the essential concepts together to move on to looking at software.

In the software world, ingredients are the so-called inputs; our cake is the output; and recipes can be found in books, which in computer science are called libraries. The recipe describes a so-called algorithm. Algorithms are step-by-step instructions on how to convert inputs into outputs. Once algorithms have been detailed out and written down so that they can be executed by a computer, they are called programmes or software. The author of the recipe is the programmer and all kitchen equipment is the so-called hardware. You, as bakers, form an additional, particularly important piece of hardware, the so-called processor, often called the CPU (central processing unit). The processor executes the individual steps of the programme, just as you follow the individual steps of a recipe.

That’s it! At least in the analogy, you have now understood what a programme does and how it runs on hardware. As an exercise, let us now transfer this basic idea before we turn to real programmes. Imagine a fully automated bakery that can bake different cakes. Before reading on, please briefly consider for yourself what constitutes inputs, outputs and hardware. Identifying programme and processor is a bit more difficult here, so before moving on, we should just make sure the concepts are clear:

• Inputs are the selection, such as the number, of a specific recipe and the corresponding ingredients.
• Outputs are the finished cakes.
• The hardware here, in addition to (large!) ovens, mixers and blenders, consists of egg beaters, conveyor belts, packaging machines and so on.
• These are all in turn individually controlled by software, special programmes that run on processors in these machines.

Now it becomes a bit more complicated. Interactions between individual hardware components must be coordinated in different ways, depending on the recipe being baked. Here, this is no longer done by you as the baker, since the entire baking process is automated. Instead, this is now done by a control computer (the processor), which again runs a special programme, namely a control program. Depending on the recipe to be baked, this programme controls the egg beater, mixer, conveyors and so on step by step in the correct sequence. In this way, it ensures that the individual stages of the bakery or baking process are supplied with the right intermediate product at the right time. Hence, the algorithms, which were originally just the recipes, are now extended in programme form to include control commands for the various machines, in other words our bakery hardware.

So, what exactly is software? Software is: programmes that run on one or several computers. Programmes such as Excel or your e-mail client are software that runs on a computer in the sense of a desktop, laptop, tablet, or smartphone. Programmes such as Google search, Facebook, or Instagram run, to some extent, on your laptop or smartphone to provide you with interactions. At the same time, their functionality, that is the actual execution of a search or provision of search results, images or messages, is made possible by an interplay of several, often thousands of computers. The existence of these computers usually remains completely hidden from us.

To a large extent, the functionality of your car is also provided by programmes. In a modern car, there are often more than 100 computers running completely different programmes. These control how you accelerate and brake, how the windshield wiper works, or how you move your seat. Software actually controls all the devices in our immediate environment. Because you can’t see these computers running the corresponding programmes and because they are usually built into devices, these are referred to as embedded systems. Furthermore, because these programmes react directly to stimuli from the physical environment and, conversely, can also influence this physical environment (temperature, distance to the car in front, and so on), but at the same time virtually exist as software in cyberspace, they are also called cyber-physical systems. Such systems can be arbitrarily large and complex, such as the ISS space station, but also very small, such as a lighting control system in a living room that simply reacts continuously to external lighting conditions. Engineers today are working to build more and more systems on a variety of very different scales. Using networks, they can then be interconnected to enable communication between systems. You have heard the idea of connecting smaller devices under the buzzword Internet of Things, which enables us, for example, to automate and network lights, stoves, heating, energy supply and irrigation in our smart homes and gardens. Clearly, this poses security and privacy problems, but that is a subject for another time.

Nowadays, by the way, engineers in interdisciplinary projects usually no longer look at these technical systems in isolation, but at so-called socio-technical systems, in which human actors are also added to the interaction of various technical systems. A good example is hybrid traffic, in which both human-controlled and autonomous vehicles exist simultaneously and must react to each other. Furthermore, because people do unexpected or “illogical” things – for instance when children paint the road with chalk – programmes must be designed to provide them with inputs that were not anticipated by programmers.

Programmes are executed on computers. Programmes and computers, that is software and hardware, behave as mind and body in humans. In addition to the processor, the hardware of computers consists of other components that perform the central tasks mentioned above. Whilst the processor performs the calculations, there is also hardware for data storage. Data is stored in memory, on the hard drive, on USB sticks and increasingly in the cloud. Communication with other computers is then achieved via connections to the network, which we will omit in this article for reasons of space. In addition to the processor as well as memory and network connection, a computer also consists of other devices, which we call hardware like the processor, memory and network connection above. These devices are: screen, fingerprint sensor, temperature sensor, battery, camera, keyboard, speakers, microphone, cables and so on. All in all, software uses this hardware to perform the tasks of computing, communicating, managing and storing data as well as interacting with the environment.

When baking a cake, inputs are the ingredients and outputs are the finished cakes. However, programmes not only receive data as inputs, but they also deliver data as outputs. Intermediate results from the individual steps are also data. There are programmes such as Internet search and your e-mail client, or the control software in your stove. You can enter commands via a keyboard, voice, or gesture, namely a search query or a message. These programmes then show you something on a screen.

In the case of a car, a washing machine, a heating system or an automated bakery, the programme also reacts to stimuli from its physical environment: the adaptive cruise control recognises cars in front of you, the washing machine reacts to water levels, the heating system reacts to temperatures and so on. The corresponding input data is provided via so-called sensors. Sensors are, for example, thermometers, cameras and radar. They translate real-world phenomena, such as temperature, cars driving in front of you, and water levels, into data that can then be processed by programmes. Unlike your e-mail client, the output of a programme in a washing machine or car is not just something you read on a display. Instead, the output data is also translated into effects in the physical environment: accelerating or braking, pumping out water, turning on gas burners and heating water. In the modern world, programmes are literally everywhere.

We have seen that recipes, algorithms and programmes describe individual steps that determine how inputs are converted into outputs. For you as bakers, these individual steps are essential. You must now consider it possible that your children, unlike yourself, are not interested in these individual steps, but only in the cake eating itself, that is, the output. Let’s further assume that your children are craving marble cake today, but the fridge is empty. They will then have to go shopping and that’s when they become interested in the ingredients, in other words: the inputs. If you want to describe the relationship between inputs and outputs, but without explaining the individual steps, this is called a function. You know functions from high school: Addition, for example, is such a function with two summands as input and the sum as output. When you add in your head, you don’t even think about how to do it, as the individual steps don’t interest you! But how does a machine do that? That is exactly the task of the programme, which describes how a function is calculated.

The connection between function, algorithm and programme is so important that I would like to summarise it once again. The function describes only the relationship between input and output, without saying exactly how the output is calculated. An algorithm defines the steps by which this function can be calculated. And the programme is the description of an algorithm in a form understandable to the machine (a “pinch of salt” is too vague to be understood by a computer!). With the distinction of “what” and “how” we can also say that the function describes the problem (what?), and that the algorithm and programme describe the solution (how?).

A programme, as with a cake recipe, is a piece of text. When you write something down, you have to find a language for it. Computer scientists have developed programming languages for this reason. A programme consists of several “sentences” in this programming language. Each sentence helps transform the input into the output, in a step-by-step manner.

There are many programming languages, just as there are many natural languages. If you know a foreign language, you may have noticed that certain concepts can be formulated more elegantly in this language than in your mother tongue and vice versa. This is exactly the reason for the multitude of programming languages. Certain technical steps in the transformation of inputs and outputs can be described better in one programming language than in another. Perhaps you can imagine that a programme that represents the interface of your e-mail client is easier to write in a language that talks about “window”, “button” and “mouse pointer” than in one that converts the distance to the car ahead into a command to brake. Have a look at the programming language Scratch which is especially good for interacting with characters in your own first computer game. In Scratch, by the way, the “sentences” are formed by pictorial symbols.