Js 2017 ― what happen, what to look out for
Last year I wrote an article about why i’m happy ― or at least not depressed:P ― to have somehow settled in web/js space. So now, a year later it’s time to revisit my post and check if I still share this mindset.tl;dr;
I’m still a fan of npm and node.js package management. Npm v4 was just released a few weeks ago and facebook released yarn which is already pretty stable and actually a nice alternative to npm (faster + ~shrinkwrap included).Node.js
Not much to say here:) Node.js and v8-engine are still evolving fast. Just take a look at the release notes. Node.js v7 now has nearly 98% support for es6 features and v8 -engine gets continuous updates/optimizations for es6 operators.Meteor
A year ago I praised Meteor and used it for basically everything ― small prototypes to real apps. Today I basically only use meteor where I don’t have a choice, but don’t get me wrong:
That’s not a “I’m leaving meteor” post.
Meteor is still as great as it was a year ago(to be true it’s even better now) ― it’s just that it can’t keep up with the fast evolving js world. I still maintain meteor projects and I don’t have more(basically none) problems like a year ago.
Arunoda Susiripala s announcement of leaving meteor was sad and I think a really hard hit for the meteor community ― I personally admire his work a lot.
That said, everything you actually needed is still there. Storybook is still supported, react-composer too. React-router-ssr was forked and is still working well. Uniforms is still maintained well. Simple-schema is working like it was a year ago. So all in all, for node.js beginners i’d actually would still recommend it as a fullstack to get started, but everyone who learned to know the benefits of e.g. webpack and hmr should probably consider to go with sth else. I’m really curious if they actually manage to release the npm transition in the near future. If yes, I thing this will produce a major boost again. Imagine sth like create-react-app, but for meteor with minimongo and co. That would actually still be top notch.Cool things to lookat:
create-react-app : already there for a while: still a pretty awesome base to get started. Currently my most used choice ― even if you have to eject right after the installation.
next.js : actually a pretty great alternative to create-react-app, coming with it’s own router and code-splitting, easy ssr support and stuff like that. I just tested it on small side projects, but it’s already pretty great… I was only really missing dynamic routes yet, but V2-beta has just been release so let’s see what’s there to come.
firebase:i was skeptical at first, but I used it in various smaller apps over the past few months and it’s actually already pretty cool and evolving fast. It’s kind of sad that you still have to leave the “ Forget about infrastructure ” comfort zone for most firebase web projects(as you most probably need ssr or at least a small server component for emails or stuff like that). Anyways i’d recommend it to everyone and am happy to see where this goes next ― even if on web it feels a bit like the unloved child in comparison to the iOS and Android version.
material-components : which is basically v2 of mdl had it’s first alpha release and i’m curious to see how it evolves and if e.g. react-mdl manages to keep up with the new version.Stuff
Of course, there’s a lot more cool stuff going on ― about most stuff i don’t know enough or nothing at all. So here are a few things I want to try in the near future… e.g. preact ― a fast, small alternative to react. graph.cool <- still not sure what i think about all that graphql stuff, but i’ll definitely give it a try. And also outside react world: The first beta of angular v4 has been released. Polymer v1.7 has been released. Browser support for webcomponents has evolved a lot. IE is next to dead (≤v10), Safari is also slowly dying(≤8):)!
Thanks for reading and happy new year:)