[nawitus]

>if I can figure it out and get lots and lots of work done with it, it's neither horrible nor overblown

That's a weird argument. Just because you think it's not horrible or overblown doesn't really mean much. That's just an opinion.

I've worked full-time with AnguarJS over a year, and I know it pretty well. It's easily better than jQuery spaghetti code, and probably better than most data-binding frameworks. It has lots of issues, though. From bad documentation and performance to the mess of $scope.$apply/digest loop and a lot of moderate design problems.

Anyway, the biggest problem is the boat. AngularJS is really trying to "eat" other libraries and actually doesn't try to be modular. A data-binding/MVC framework shouldn't invent it's own module system, and recently it's own programming language. Apparently Google is trying to kill of jQuery too and substitute their own system. And they want to reinvent everything as AngularJS modules instead of "plain" JavaScript code that can be used across multiple frameworks.

What I'd like to see in AngularJS 2 is a tiny core which does data-binding without $scope.$apply, is fast and modular and could be used with a wide range of other frameworks. What I don't want to do is start writing AngularJS modules with AtScript and using their forced(?) ES6 classes.

[Bahamut]

Angular 2 definitely aims to be more modular - the Angular team acknowledges the problem that it isn't modular enough itself. Stuff like di.js, watchtower.js, zone.js, etc. are all built with modularity in mind & are framework agnostic. I think once Angular hits 2.0 and it becomes highly modular, we are going to see a boom in development productivity. Some of what is currently in Angular is unique for JavaScript, such as the change listener pattern and dependency injection. These are things that we'd like to take advantage of with other libraries. Combining stuff like di.js with React sounds very appetizing to me.

In addition, it would appear that we would be able to use these libraries with Node.js as well.

I can only see this being a good thing.

[Jgrubb]

> That's a weird argument. Just because you think it's not horrible or overblown doesn't really mean much. That's just an opinion.

A contrary opinion to the OP's opinion. Are you asking me "where's the data" (?), because this whole thread is opinions.

edit:

> It has lots of issues, though.

No argument here. Providing validation feedback on forms, for instance. I'm not arguing that it's perfect and that I'll never love again, but it has opened up my head this year and for that I'll go to the mat for it.