Mozilla should get more ambitious with Rust

As the programming language hits version 1.5, Mozilla should build both with it and on top of it

Mozilla should get more ambitious with Rust
flickr/Till Krech (Creative Commons BY or BY-SA)

Mozilla's low-level language Rust, a secure and speedy competitor to C/C++ in both applications and systems programming, unveiled its 1.5 version yesterday. The news comes hot on the heels of Mozilla canceling further work on its Firefox OS project -- a sign that Mozilla's time and effort may be better spent building groundbreaking new ecosystems instead of trying to copy existing ones.

The big update for Rust 1.5 is a change to the Cargo application-packaging system. Typing cargo install allows apps hosted on to be installed on the local system. It's similar to how the pip install command is used in Python to fetch and install libraries from the official Python repositories.

This allows applications to add subcommands to Cargo, some of which are already available. For example, there's the ability to statically check a Rust project without building it (cargo check) or build dependency graphs for the code (cargo graph).

Other improvements include less memory use during compilation, a smaller core library, and better performance in builds for unoptimized programs. That said, some of the truly sophisticated planned features for Rust, such as incremental recompilation, still haven't been given a target date yet.

Rust has become one of Mozilla's star initiatives. The Servo rendering engine, built in Rust and designed to exploit multicore systems for faster page rendering, is set to eventually land in Firefox as a way to accelerate its performance. No details are available yet about the time frame, but the first support components have already been added to Firefox proper.

If Rust becomes widely adopted, Mozilla could further benefit from its popularity by offering enterprise-level services around it. Hosted build services, application telemetry, or other software development features enjoyed by enterprise developers are all possibilities.

Since Rust aims to be a systems language and not just an application language, this could include things like automated, cloud-hosted build-and-test functionality for applications that normally run on bare metal. Not only would it provide Mozilla with a bevy of ways to get telemetry about how Rust is being used in the wild, it could also provide a revenue stream. Maybe not one as sizable as its search engine cash, but easily far less problematic.

Right now, though, Mozilla's pitch for Rust, as described in a recent blog post, remains modest, and mainly revolves around how it's being used to accelerate Mozilla's products, since "the future of the Web requires high performance and security."

Copyright © 2015 IDG Communications, Inc.