[zackbrown]

I remember sitting in the front row at ngEurope (Angular conf) back in 2014, when Misko announced that Angular 2 would be a (breaking) API rewrite vs Angular 1. At that time, Angular was perhaps at its global maximum of community mind share—bigger than React!—and its popularity began to sink afterwards in large part due to that PR snafu.

I wonder if this is React's "Angular 2." What's worse, the misstep this time is due to ham-fisted corporate mal- stewardship, not just miscommunicated technical good intentions. Vue is similarly positioned to be the "next thing" in the way React was three years ago. My how this space churns.

[rogihee]

I'm still working heavily with AngularJS (1) and in combination with components (1.5+), TypeScript and a MVVM approach it is very clean and productive. Even for all its flaws, it still very easy to start with in 10mins, include 1 lib in 1 line and "Hello World" is up and running.

Compare that to the space ship Angular 2. They really made an outstanding effort to create a huge barrier of entry for anyone not having a degree in DevOps. I'm an old fart, I don't need your stinking CLI, Grunt, Gulp, NPM etc stuff.

Compare that to Vue.js really shines in this regard with the same level of entry as AngularJS. For me that is part of what I thinks makes it a succes.

When choosing a new framework I tend to not look for the future, but look in the past. I chose what was mainstream 2-3 years ago, which means right now still AngularJS for me. This means I'm a leecher on StackOverflow, but so be it. I have my own company and clients pay me for working applications, which means I have to choose my tools as efficient and effective as possible. By choosing last years model, it is less sexy but will get me from A to B with the same speed plus added reliability. At the same time I'm keeping a close look to Vue.js and if still going in the same direction next year as it is now, I will definitely switch to it.

[Bahamut]

Coming from the Angular community & still connected quite a bit to it despite currently working primarily with React, I don’t think having an incremental change would have helped Angular compared to React due to the complexity of Angular.js’s API & its flaws. I actually stand by quite strongly behind the decision to rewrite in Angular 2, but Angular 2 took a long time to come into being - it was rewritten 3+ times in getting to its current state & 3-4 years in the making. I think Google devoted the proper resources too late into it in its quest to make the most performant & flexible framework. I believe this gap is what largely hurt Angular, and while the PR snafu was visible, I think the current situation would have been arrived at regardless.

[esistgut]

I agree with you and I think Google is still underestimating the amount of required resources because the framework is still not really "done", a couple examples: universal (SSR) just hit a stable release on angular-cli so it needs time to be widespread and supported in the ecosystem. The are core components like the i18n waiting for a major refactor because they lack basic features: https://github.com/angular/angular/issues/11405