Thoughts about Deno? Will it replace node.js?

maybe over the time of 2-3 years

Not in the short perspective, node is to big.

If it proves its better it might over time.

Have not checked if its a drop in replacement or if you need to make any changes to use it.

If it requires changes it will likely take much more time.

Fancy pants features don't cause replacements. Some of the largest companies in the world have invested their monies in node. It takes politics to change it, not programming.

@Voxera Definetely not a short term drop in replacement. Besides not using NPM, all dependencies also need to be ESModules.

Nah fuck no.
Most people won't switch their working nodejs applications to something just because it's new and neat.

Absolutely not.

I think some new applications that have no concern for risk will use it. But server software is usually expected to have a great reputation before it is widely used. Deno is in version 1. One. It's gonna be a long time before it's actually seen as trustworthy, battle-tested, and viable enough for companies to invest their existence on. Because for some companies, a bad web server means death.

I think in the long run it will be like MVC. Clearly better than ASP, and even uses the same language, but ASP is still around. Even worse, there's a billion packages for node and I don't even know what deno has, but unless deno supports node packages, you're losing a lot of development time trying to find or even needing to create shit like .mp3 duration packages. There's a lot of very specific problems in programming which is why npm is so huge. It had an answer for everything.

Deno is nice. But don't forget that node has support for es6 modules, too. And they have support for requirejs, obviously. And you can even use both at the same time.

Node is in version, well, last time I checked it was 13. Fucking 13. That's battle tested. It's a household name.

Deno is in a critical stage where either it will explode or it will fade away to a GitHub repo that gets two or three new commits a year

Anyways, final thought is no company with any understanding of risk will adopt it today. But tomorrow (metaphorically)? It's possible for sure!

This twitter thread might be helpful.

https://twitter.com/eaf4/status/...

If anything some ideas might get blended with Node from Deno just making node better. I honestly don't see it as a replacement. Node is battle tested, has a fuckload of packages and tools and the ECMA standard keeps improving the quirks around the language.

I feel deno is like android permission system where you need to give permission for app to make something usefull. It will make automated stuff very hard to run.