Hacker News Reader

Comment by jFriedensreich | Hacker News Reader

Everyone should take deno seriously, at this point imho it is pretty clear that nodejs is legacy and deno the future. It will not happen tomorrow of course as especially tooling and the nodejs compatibility layer has a long way to go but never dealing with node_modules again and a sound security model as well as web apis instead of custom nodejs ones is allready enough to make it killer, on top there is the integrated typescript runtime of course...

View on HN

Thanks for your comment, and I'm glad to see that you are excited about it! I personally won't be an early adopter and I don't want to do any of the work, but it would be great if lots of people get on board. I'm personally really happy with the current state of JS tooling, and I think node_modules are generally great and work fine. So I'm not too convinced about Rome or deno.

I'm personally looking forward to building a new toy project with Svelte, Sapper, Tailwind CSS, and Snowpack (with TypeScript, Node.JS, and NPM.) This still feels like the future to me.

As an aside, Sapper in particular is probably not the future, as there’s been an unofficial announcement[1] that it will be replaced with file-based routing built in to the Svelte app template.

1: https://youtu.be/qSfdtmcZ4d0

That's totally fine and happy you have something you feel at home with, not everyone needs to be an early adopter and i would never advice people to jump on something when they think it might be a hype train but only when they see real advantage. The js ecosystem is famously evolving too fast to keep on top all the time anyways, we should also be allowed to relax and refine. By the way i think no one expects you to migrate your existing healthy projects to deno currently anyways. The current adoption cycle is at adopting js libraries like svelte to support js modules and deno, switching to deno for one shot scripts and new command line tools, developing replacements for basic nodejs libraries that dont make sense in deno or seem too archaic and building smaller internal tools and apps in deno and while doing so evolve the workflows, tools and best practices every step in the way. Especially importmap support and workflows need some evolving...

Reply on news.ycombinator.com