Deno is now done!

I was a Deno enjoyer - it was a serious project with an amazing runtime, fully integrated with TypeScript. Not only that: it had everything you needed to build (outstanding standard library, linter, formatter, test engine), deploy (Deno Deploy, KV, queues), and distribute (bundler, compiler, JSR) your code.

It offered a safer way to use TypeScript - secure by default, with no file or network access unless you grant it - requiring less configuration, a better DX, and no support for node_modules when it started 🙏.

Check this video for an overview.

What really happened?

Cloudflare hired the Deno team to build their internal runtime (which gives me hope). Cloudflare is a great cloud company - my default choice for building for the web - and they open-sourced their runtime to counter the claim that Cloudflare is a huge lock-in you’d never be able to leave, even if you wished.

Don’t get me wrong, I’m a paying and happy customer of Cloudflare, but I do consider Cloudflare a huge lock-in - we all have to lock in to something, am I right?

Even better if my cloud provider at least makes some effort to open-source their work - yes, AWS, I’m looking at you 👀.

Celld

Celld is an open-source implementation of Cloudflare Durable Objects - a promising one - which seems like a great primitive to build on top of.

Not much to say - I haven’t tested it yet, but I’ll give it a try eventually. For Cloudflare, this was a big deal, probably the biggest reason for that decision.

Being honest

Deno is not actually dying - it will continue to receive support for some time, but the core team will be focused on Cloudflare.

The community can pick it up, fork it, make it better - but we have to be honest with each other: losing the core team makes things harder!

As for me, I’ll keep building on Deno for now - but for anything new, I’m taking a hard look at Node, Bun, and Workers.