With numbers, we can create other types of data. One of the most common types of data using a computer is written text. Text can be converted into a sequence of numbers by using a code where each letter corresponds to a number. For example, A is 65 and G is 71. You could just as easily say that each letter corresponds to a bit pattern. A is 01000001, but I find it easier to think in numbers. The computer doesn't carry the weight. The American Standard Code for Information Interchange abbreviated to ASCII is a standard for representing texts that includes all of the letters in the English alphabet plus numbers and special characters in seven bits. That's great if you're writing in English. But what if you speak in another language? An extended eight-bit version of ASCII includes the accents for European languages like French or German. But that still doesn't help you if you want to represent Arabic, Bengali or Mandarin on a computer. Since that excludes the vast majority of the world's population, ASCII is being replaced by a new standard called Unicode which can represent almost any writing system and, importantly, is extensible so other writing systems can be added. It's an unfortunate legacy of the Western bias in computing technology that there's still a lot of computer systems that use ASCII but that's gradually fading. A very large proportion of the data represented on a computer is a combination of text and numbers. For example, a web page is mostly text with some numbers used for formatting, most spreadsheets and numbers with some text. A corporate database will have texts for things like employee names and numbers for salaries and payroll numbers. So, text and numbers can get us a long way. But what about other things? What about images and sounds? We've seen how we can make pictures out of black and white squares represented as bits. We call each square a pixel, a picture element. A modern picture is typically made out of millions of pixels. For example, an image that's a 1,024 pixels across by 1,024 pixels down would be over a million pixels in total, but it's still relatively low-resolution. Of course, we don't just want black and white pixels. The first improvement is to have varying shades of gray or brightness between black and white. We could do that by using numbers instead of a single bit. Because computer screens create colors using light, high numbers values represent bright colors, more light, and low number values represent dark, less light. Brightness is typically represented by eight bits. So, shades between black and white will be a number between 0 and 255. What about color? As we saw earlier, to create color, we use three numbers, one for each primary color, red, green and blue often called RGB. By combining them, you can create any color that is visible to humans. We often add a fourth number to allow for transparent images. This number normally called alpha represents opacity, which is the opposite of transparency. So, an alpha of 0 means a completely transparent pixel and 255 is completely opaque, not transparent. Once we have images, we can create videos by having sequences of images. Each obstruction allows us to create new obstructions. Numbers lets us create images which lets us create videos. Again, this is a concept we've already discussed in this course. Sound is also represented in terms of numbers. A sound is actually a series of microscopic changes in air pressure that is picked up by our ears. A microphone can record these changes in air pressure and convert them into voltages. To record audio in digital form, we can record the air pressure change as a number. To get realistic sound, we have to do this very fast, typically, over 44,000 times a second. One of these individual recordings is called a sample and a whole sequence of samples becomes a sound file. So, our very simple bits can be used to build up very complex data. A single number is tiny, only 32 or 64 bits. But images and audio can include millions of numbers. Since a video file is a long sequence of images, it can be very large, often several gigabytes. Our large memory chips and hard disks can hold what seems like huge amounts of numbers, but with data like video, they can often fill up. So, it's very important to be aware of how data is being represented as it can directly affect the performance of your computer. You've now seen how bits as simple components can be built into complex data. This will help you understand existing data representations. But it's also vital to one of the most important jobs that you will do as a computer scientist: designing your own data representations.
