Ask HN: Why is everything in JavaScript changing so fast?

Because nobody has yet invented a compelling way to write JavaScript applications, so people keep trying. We're still waiting for the Ruby on Rails of frontend development.

I'm not saying Ruby on Rails is the best backend framework (I don't use it anymore), but when it came around everybody understood that it was getting something right. It popularized a new model for server-side web development that all other communities started to follow and improve. When it came out, most developers who were already working with the web could see that it would be very useful, it solved real-world problems they had, it just made sense.

This is much less clear with JavaScript frameworks. They usually have a tough learning curve and the benefits are not always coming as expected. So people keep trying to find a model that works well.

I would say react is the way to write large data driven applications, but JavaScript, without a tool like Google closure, is not suitable for 100k lines of source code app that is being actively developed. Personally I am playing around with typescript and webstorm and that seems to me to be good enough.

I think the other poster meant 3rd party libraries, rather than the runtime.

That said, the docs on (most of) the libraries I have used are generally pretty good.

* Prototype - excellent docs (but library since overshadowed by jQuery)

* jQuery - initially the docs were hard to follow, but they seem to have gotten better

* Angular - good enough docs, and plenty of books on the larger concepts of how things fit together

* moment.js - excellent docs

* validatejs - excellent docs

* Ramdajs - generally excellent docs, but I am less familiar with some of the FP concepts used in a few cases.

The quality on some of the jQuery and Angular plugins varies a bit, admittedly.

I remember the, well, "rage" I had at Sun when I stumbled across RoR. I had completed my Java web certification the year before. The Java stuff seemed clumsy and repetitive. One year later, saw Rails. "Why couldn't you geniuses have done something like that, instead of this mess you gave us?!?"

My theory, which I think goes fairly against what others will say here, is that JavaScript/Node is the first truly post-internet environments to reach critical mass. By post-internet I mean they consider sharing and open source as foundational to the environment (as opposed to tacked on later, or merely supported).

The "JavaScript standard library" (or lack thereof) is a clear example of this: NPM is the LARGEST collection of packages of any language, and you can almost always find someone who has already written what you need most the time. It never ceases to surprise me. During the Pokemon Go craze I was curious to play with the protocol: there was already a package to do it. I wanted to ping my find my iPhone, there's already a package to do that. Its incredible. And yet the actual built in standard library is notoriously lacking. Although not by design, certainly the end result is that that has become "OK" thanks to NPM.

Compare this to something like AppKit (and UIKit), which were very much in the pre-internet/sharing mentality: there is one framework everyone has and becomes experts in and is fairy stable, because its kind of the only thing you can rely on being around. Sharing code is very difficult, to the point where just copying and pasting files is still a valid contender.

The explosion of options and the increase in churn is a natural consequence of this. There are simply more people working on more problems, all encouraged by an environment where sharing code is a fundamental property. If step 1 of setting up node is downloading other people's random packages then it should be no surprise that step 4 of sharing your own packages will seem natural. We should expect this to only get faster -- the more people program, the more crazy ideas will be tried, the more switching there will be. I really don't understand what the alternative is: arbitrarily choosing frameworks to be around for a 5 years (a decade?).

We are simply taking the last measure of stability, which was largely governed by how long it used to take information to disseminate to the entire community, how long it used to take one controlling body to get their large waterfall releases out the door, and using that as an arbitrary comparison point to how long it takes hundred of individuals to release their individual takes on software. Asking to "stop the churn" is a strange request: there's no single entity pushing stuff out "too fast", this is the aggregate affect of all programmers in the node community publishing their ideas. Who are we asking to slow down exactly? I can understand saying the single case of Angular 1/2, but the "JavaScript fatigue" people actually feel is due to the interaction of all the JavaScript projects that each gain and lose popularity.

Unless you believe we're really close to figuring this whole "software development" thing out (whatever that means), I wholeheartedly believe that programming 20 years from now will look more like this -- faster pace of developments.