Why Google’s Carbon Language Claims to Be the Successor to C++ – What You Need to Know

At the CPP North C++ conference, Google announced a new programming language called Carbon , describing it as the "successor to C++". The language is open‑source on GitHub, still experimental, and quickly rose to the top of the GitHub trending list, gathering over 21.3k stars.

When asked for his view, Bjarne Stroustrup, the "father of C++", said he welcomes experimentation with programming languages and styles, but noted that Carbon is too new and under‑specified for a meaningful technical comment, and that providing a true alternative to C++ without new language rules, libraries, and management is difficult.

Google engineers explain why they are not simply extending C++: the language is now 40 years old, burdened with technical debt, and its focus on backward compatibility leads to massive codebases. Compilers include both open‑source (e.g., gcc) and closed‑source versions that evolve slowly, while the standards committee’s bureaucracy hampers adding new features.

The proposed solution is to avoid inheriting C/C++ legacy and start from language fundamentals such as modern generic systems, modular code organization, and a consistent, simple syntax.

Exploring What C++ Can’t Do

Chandler Carruth, the lead engineer of Carbon, says the language is not meant to compete with future C++ versions but to explore capabilities that C++ cannot provide, similar to how TypeScript extends JavaScript.

Carbon’s current highlights include:

Introducer keyword and simple syntax

Function input parameters as read‑only values

Pointers offering indirect access and variants

Expression‑named types

Importing APIs via package names

Powerful, definition‑checked generics

…

Future development goals focus on performance‑critical software, language evolution, readable and maintainable code, practical safety and testing mechanisms, fast and scalable development, modern OS platforms and hardware, and seamless interoperability and migration with existing C++ code.

The team also plans to create a built‑in package manager, a feature long missing from C++.

A simple example compares C++ and Carbon code for calculating the sum of the areas of two circles (radius 1 and 2). The visual comparison shows Carbon’s more concise syntax.

Another diagram demonstrates Carbon’s bidirectional interoperability with C++, allowing programmers to call C++ libraries from Carbon and vice versa without extra overhead.

Regarding safety, Carbon does not initially provide a memory‑safety system; the project prioritizes migration over safety, hoping to achieve safety comparable to Go or Swift in the future.

The article also compares Carbon with Rust, another language touted as a C++ successor that emphasizes memory safety. While Rust offers strong safety guarantees, its interoperability with C++ is limited, making migration of large C++ codebases difficult. Carbon is designed to seamlessly connect with existing C++ libraries, making it attractive for developers with extensive C++ codebases.

Community reactions are mixed: some praise Carbon’s potential to simplify language and improve safety, while others doubt Google’s commitment to long‑term innovation or view Carbon as redundant given Rust and D language capabilities.

Readers are invited to share their opinions on Carbon’s prospects as a C++ successor.