How do people write code? They write instructions, and they have to write them very carefully, because a computer program is a list of instructions for a computer and a computer can't think for itself, so people have to give it those instructions. The people who write them are called programmers, and the code they write is what makes computer programs. Computational thinking is the thought process of formulating problems and their solutions so that an information-processing agent, such as a computer, can carry them out. To be carried out by a computer, a solution must be a step-by-step process, which is an algorithm. Computational thinking could be described as thinking like a computer scientist, and it is an important skill for everyone to learn, whether or not they want to be a computer scientist.

Computer science is the study of computers and computing, including algorithms, hardware, software and how they are used to process information. Software is the set of instructions that tells a computer what to do, as opposed to hardware, the physical parts, and the two main types are system software and application software. Computer programming is composing and organizing a collection of instructions in a language the computer understands. Programs can be as different as scientific calculators, databases and animation software, and a programming language is what lets people write the detailed instructions for them.

An algorithm is a systematic procedure that produces the answer to a problem in a finite number of steps, which means that it is a set of steps that ends. Euclid's Elements, published about 300 BCE, contained an algorithm for finding the greatest common divisor of two numbers. An algorithm describes a way to solve a problem so that it can be solved over and over, by people or by machines. In plain words, an algorithm is a plan that people or machines can follow, and it can be used again and again. Computer scientists compare how efficient algorithms are using algorithmic complexity, also called Big O notation.

Let us start at the bottom. Underneath every program is binary code, in which each digit is either off or on, written as 0 or 1. Each decimal digit from 0 to 9 can be shown with four binary digits, which are called bits. Claude Shannon identified the bit as a fundamental unit of information and, coincidentally, the basic unit of computation. ASCII lets machines from different makers exchange data, and its code had 128 unique strings of ones and zeros, each standing for a letter, a number, a punctuation mark, a symbol or a function. Machine language is made of the numeric codes that a computer can run directly, and it is also called binary code or machine code. Machine and assembly languages are low-level, so a programmer must manage a computer's own features of storing and handling data, while high-level languages shield programmers from that work. High-level languages are designed for people to understand, and they are translated before a machine can run them. A compiler translates a whole program written in a higher-level language such as C++, Rust or Java into binary or machine code that can run.

Here are the steps of one person's story, in the order of the dates. Grace Hopper was an American mathematician and a rear admiral in the U.S. Navy, who helped develop computer technology, including UNIVAC I, the first commercial electronic computer. In 1946 she wrote the first computer manual, which described how to operate the Mark I calculator at Harvard and was the first extensive treatment of how to program a computer. In 1949 she joined the Eckert-Mauchly Computer Corp, and there she designed one of the first compilers and coined the word compiler. She made an early tool that let people give a computer instructions in English-like words instead of numbers. In 1957 her division developed Flow-Matic, the first English-language data-processing compiler, and it had many features that inspired the development of COBOL.

Before FORTRAN, programs were written in machine or assembly language, in binary or hexadecimal arithmetic, and FORTRAN shortened that process. John Backus finished Speedcode for IBM's first large-scale scientific computer, the IBM 701, and although Speedcode needed scarce memory, it greatly reduced the time needed to write a program. FORTRAN took three years to develop, led by Backus with a team of 10 IBM employees, and it combined a form of English shorthand with algebraic equations. FORTRAN became the most often used language for scientific and technical computing over the following decades, and the Computer History Museum says it is still in use as of November 2026. COBOL grew out of the 1959 Conference on Data Systems Languages. It was designed to be portable, running with little change on computers from different makers, and easy to read like ordinary English. COBOL's popularity declined beginning in the 1990s, but many large organizations, such as banks and government agencies, still used it in the early 21st century. A 1997 study estimated that over 200 billion lines of COBOL code were still in existence, accounting for 80% of all business software code.

Programs keep track of information in variables, and that is where we begin. A variable is like a box that holds a value, and the value inside can change. Variables are not the values themselves, because they are containers for values, and they can hold nearly anything, even whole functions. Every variable gets its own name, so the program can find what is stored in it, and a program can use a variable to keep track of something that changes, like a score or a count of clicks. A data structure is a particular way of organizing data so that it can be used efficiently. An array is an ordered collection of data, and each item has a number called an index, and in JavaScript the numbering starts at zero. A stack is a data structure where the last item added is the first retrieved, which is called last-in, first-out.

Here are a few more ideas, one after the other. In programming, recursion is when a function calls itself, and it is used to solve problems made of smaller sub-problems, and it needs a base case that ends it and a recursive case that continues it. In programming, abstraction reduces complexity by hiding the technical detail of a system behind a simpler interface, called an API. It can also avoid duplicate code and make code easier to reuse. An API is a set of features and rules inside a software program that lets other software interact with it, as opposed to a human user interface, and it works like a contract between the program that offers it and other software or hardware.

Here is the story of Python and of its name. Python was developed by Guido van Rossum in 1991, named after the British television series Monty Python's Flying Circus, and publicly released in 1994. You may have heard that it is named after a snake, but in fact it is named after that television series. Here is how its versions went. Python is a high-level, object-oriented language that emphasizes readability, which lowers the cost of maintaining programs, and it supports modules and packages, which encourage splitting a program into parts and reusing code.

Here is more about the way people share what they make. Python is open source, so anyone can use it for free. Open source does not just mean access to the source code. The Open Source Definition says the program must include source code and allow distribution in source form as well as compiled form. Under the definition, the license must allow modifications and derived works, and must allow them to be distributed under the same terms as the original software. The license shall not restrict anyone from selling or giving away the software as part of a larger software distribution. The Free Software Foundation's GNU Project says free software is software that respects users' freedom and community. You may have heard that free software always means software that costs no money, but the GNU Project calls this a matter of liberty, not price.

Programmers also need a way to keep track of their changes, and Git is a version control system. The Git book explains that version control records changes to files over time so you can recall specific versions later. A version control system lets you revert files or a whole project to a previous state, compare changes over time, and see who last modified something that might be causing a problem and when an issue was introduced. Git began in 2005. Its goals included speed, a simple design and being fully distributed, and it was built to work on the Linux kernel, so it could handle repositories with tens of millions of lines of code from the start.

Here is one more tool. Scratch is a coding tool where kids make their own stories, games and animations, and as of November 2026 it is one example of a tool built on blocks. It was developed at the MIT Media Lab and released to the public in 2007. It is built on blocks that kids drag and snap together, so there is no tricky syntax to type, which is how the Scratch Foundation describes it as of November 2026. In 2019 Scratch was transferred from MIT to the Scratch Foundation, an independent nonprofit that develops and maintains the Scratch platform and community. In short, it is a tool that you use by snapping blocks together. As of November 2026, the Scratch Foundation says Scratch is free because people make it possible, and that its moderators work to keep conversations and projects safe, kind and welcoming.

Software developers design computer applications or programs, and the U.S. Bureau of Labor Statistics groups them with quality assurance analysts and testers, who identify problems with applications or programs and report defects. As of November 2026, the bureau reported a median pay of $134,040 per year in 2025 for software developers, quality assurance analysts and testers together, and 1,905,400 jobs in 2025. It gave the median annual wage in May 2025 as $135,980 for software developers and $104,300 for software quality assurance analysts and testers separately. These workers typically need a bachelor's degree in computer and information technology or a related field, and some employers prefer developers with a master's degree.

The bureau projected that employment of software developers, quality assurance analysts and testers would grow 10 percent from 2025 to 2035, much faster than the average for all occupations. About 106,100 openings a year are projected for these workers. Many of the openings are expected to come from replacing workers who move to other jobs or leave the labor force, such as to retire. All of these figures are as of November 2026, and they are the bureau's own projections.

Software also has to be made safe. NIST's Secure Software Development Framework, SP 800-218 version 1.1, gives recommendations for mitigating the risk of software vulnerabilities, as of November 2026. The U.S. cybersecurity agency CISA and international partners published Secure by Design guidance, first in April 2023, urging software makers to ship products that are secure by design.

Mistakes are part of the work, and bugs come in two main kinds, which are syntax errors, mistakes in writing the code, and logic errors, where the instructions are wrong or in the wrong order. Programs are checked by testing. Testers and programmers test constantly, and tracking down a bug takes logical thinking to work out where and why something in the program is going wrong. Debugging involves testing the outcome of programs and looking for any errors or mistakes, either in the original algorithm or in the final program code. Programmers can find a bug by checking a program one small step at a time, and a source-level debugger lets programmers look at variables, set breakpoints and step through the code a line at a time. We can test that a program works by giving it lots of different inputs. It is okay to make mistakes in a program, because finding and fixing bugs is a good way to learn.

Words to know

• abstraction: hiding detail
• compiler: code translator
• recursion: self-calling function
• vulnerabilities: security weaknesses
• syntax: writing rules