Mojo went open-source on a Tuesday. The systems-language community is louder than usual.

Modular pushed Mojo to GitHub on Tuesday, April 30, under an Apache 2.0 license. The repository imports cleanly. The build works on the three reference platforms (macOS arm64, Linux x86_64, Linux arm64). The accompanying RFC archive — 117 documents covering five years of language design decisions — is more interesting than the language itself, and it is what I want to write about.

The argument I will make is that programming languages, as cultural artifacts, are most clearly seen not in their syntax but in their decision logs. The Mojo RFC archive is one of the most generous decision logs a systems language has published since Swift's — and Swift's, recall, was published over many years rather than in a single drop.

What the RFCs say about the language

Two RFCs in particular are worth reading. RFC-0014, "Reference Semantics by Default," documents the team's decision in 2022 to lean toward Rust-style borrow checking rather than the value-semantics-with-lifetimes model that an early prototype tried. The thread runs to 38 comments. Three of them are from Chris Lattner directly, in the kind of measured tone Lattner used to use on the Swift evolution lists. RFC-0042, "Concurrent Memory Model," is the one that documents the decision to ship without async coroutines in v1.0 and adopt them as a v1.1 addition. The thread there runs to 71 comments and includes a paragraph-long contribution from a former Rust core team member.

A programming language is a cultural artifact. The RFC archive is the only place the culture is fully visible.

What this is, in the longer arc

Adopting Mojo is, in 2026, the kind of decision a senior engineer makes for a small team or an academic group; it is not yet the decision a Series-B startup makes for a five-year codebase. That second decision will hinge on the package ecosystem, which the open-source release is meant to seed. The interesting comparison is not Mojo vs. Rust, which is the headline-friendly version. The interesting comparison is Mojo vs. Zig, which is the actual sociological question. Both languages are now in the "we are open and we want a community" phase. One of them is going to win the next two years of academic adoption. My guess — and it is a guess — is that the language with the longer-running RFC archive has been thinking about the people problem longer than the syntax problem, and that pays off.