Why is server-side WebAssembly exciting, and why now?
WebAssembly beyond the browser is one of the most exciting trends that’s still in the early days. I’ve been thinking about the market tailwinds that could push it forward (all referring to server-side Wasm only). Here are some of them:
- Edge computing is the next major platform shift, and developers are looking to build applications in a platform-independent way. Wasm’s portability, low resource consumption, and lack of a cold start problem make it ideal for the edge.
- Rust is becoming one of the de facto languages for distributed computing. Rust is a popular language that compiles to Wasm, and the future of Rust & Wasm seem closely linked.
- The “competition” between Wasm and Docker is false. All technologies have trade-offs, and every piece of infrastructure is better for some things over others. There are big markets for both Wasm and Docker.
- WASI is here, and more people are building the developer toolchain for Wasm. This creates a virtuous cycle. Better developer tools = more Wasm usage. More Wasm usage = more incentive to create better developer tools.
- I see an opportunity for startups to build Wasm-specific developer tools — including application frameworks, registries, build tools, testing, debugging, orchestration, and observability. Some companies are already working on this, and many existing tools still work for Wasm.
- Wasm is fast, and performance often creates winners in markets. I’m curious if ecommerce sites of the future will demand search that’s written in Rust, compiled to Wasm, and served on the edge.
- More people want to add functionality to an existing system (e.g. platform extensibility). Wasm’s sandboxed modules require you to explicitly grant capabilities, which is safer for untrusted 3rd party code. Wasm gives developers more language options to write these extensions.
- Supply chain attacks are increasing almost exponentially. Packages have dependencies that create security vulnerabilities. Several infosec companies are working to secure the supply chain, and the Wasm community could contribute to this goal given Wasm’s security benefits.
- Talent goes a long way, and this shouldn’t be underestimated. I’ve been impressed by the many talented people that are excited about Wasm and working towards its future. The growth within the @bytecodeallies is one example of this.
- I’ve also talked to plenty of naysayers, who all make valid points: Isn’t this market very early? Don’t several key languages have only experimental support for Wasm? Doesn’t Wasm make more sense for the browser? Is this another JVM?
- Perhaps most importantly, I’m keen on ideas that could change industries versus focusing on the (many) things that could go wrong. And as always, I’m still learning! I’m interested in other points of view, so please reach out!
This post was originally written on Twitter. If you’d like to chat about sever-side Wasm, please direct message me on Twitter (@reneeshah123). Also here is a link to the original post: https://twitter.com/reneeshah123/status/1392208269628743680.