Presentation: From Rocks to Rust: Our C to Rust Paradigm Shift

We're working on bringing to cloud an existing project for one of our clients. It involves a lot of work with products that are quite old, 20 years old, that were written in C, and we're turning a lot of these old products into microservices, containerising them, looking to get the whole system up and working in the cloud for our client, which involves a lot of cool interfacing with new languages like Rust, trying to build these wrappers on our own products to enable easier RESTful HTTP interfaces. It's quite an exciting mix of the old stuff that we used to do with a lot of new groundbreaking stuff that we're trying to be pioneers for.

We want to evangelize Rust as a company. We made the decision to move to Rust as our preferred programming language going forward. So it's in our best interest that people do the same thing. But we’re not just motivated for own gain! - we genuinely do think that Rust is a great language and that we've got a lot of benefits from using it. We want other people to be able to take advantage of these benefits as well.

I think that's a very legitimate concern and it is something that I directly address in my presentation. The short answer is, yes, Rust has a learning curve. This is something the community has acknowledged and is already taking steps to mitigate. If you pick up Rust now compared to a year ago it's already much simpler and easier to get to grips with. In terms of Go being simpler and perhaps more lightweight and easier to get started with, yes, I think that might be true but there are limitations to Go. A notable one is that Go needs a garbage collector whereas Rust doesn't have that. For the extremely fast high-level processing that we need to do as part of telecoms, Rust was definitely a better fit for us. And there are other factors too that I address.