5 ms·

The problem I see with this is that it's very different from Angular 1 to the point people using Angular would have to learn it all over again. Breaking how th

by inglor 12y ago

The problem I see with this is that it's very different from Angular 1 to the point people using Angular would have to learn it all over again. Breaking how the framework looks this drastically is both very brave and very risky. Especially since from discussions with Angular core (and as you can see from the presentation) - Angular 2.0 relies heavily on ES6 features and syntax which are not yet supported in any browser and worse - are not known by most web developers anyway.

Spiritus 12y ago
I don't know, somebody has to be first though. Also, from what I've heard it's like two years away anyway.
tracker1 12y ago
I'm only guessing that it'll be another year or so before Angular 2 is in a release-ready state... It's a huge break from the prior version, and imho a good one. The tooling will be the most interesting to me, I can only guess that they'll use node and traceur for building the output in a project to something a browser can use. I've been moving towards React (and Flux) as it lines up well with the way I've been developing with node. This just takes it a step farther. As for the features that aren't well known, this will change and have to be learned. There's enough in the ES6 stuff that I'm running new development against Node 0.11 so that I can get generator support and some other language features in place. Though still using the CommonJS/Node style require() statements. I do hope, that like React and its' tooling that the Angular guys have a compelling case and support for shared server/client logic and rendering as well as support for Browserify and Webpack+hot-reload. Though I like Webpack I think it goes too far from the node system, it is really nice however and I can see why others use it. It's all really interesting... I'm curious to see where Polymer goes from here as well...
cageface 12y ago
Agreed. This looks almost like a complete rewrite, which seems like a good thing because IMO Angular 1 is overly complex. But it does make me wonder if I should bother to really learn Angular 1 at all. The main thing pushing me towards Angular has been the size of the community but if they're wiping the slate then I might just go with React after all.
- tracker1 12y ago
  Angular has some huge support and momentum behind it... there has also been a lot of job postings around it as well. That said, I couldn't get past some of my issues with it, and that server rendering (shared code) was not really a good option there... The issues with dependency injection is another issue, and testing pretty much requires a browser. These are the reasons I'm using flux/react now. I'm really looking forward to following the development of Angular 2 now, and may just be my tool of choice for new development in a couple years.
nawitus 12y ago
Google is developing AtScript, which compiles down to ES5. I don't think that's the right way to go, though..
dragonwriter 12y ago
> Angular 2.0 relies heavily on ES6 features and syntax which are not yet supported in any browser Which is why part of the project is a compiler which compiles ES6 (actually, an ES6 superset) to ES5, which is supported by every browser. https://github.com/google/traceur-compiler https://github.com/google/traceur-compiler

by inglor 12y ago

Spiritus 12y ago

I don't know, somebody has to be first though. Also, from what I've heard it's like two years away anyway.

tracker1 12y ago

I'm only guessing that it'll be another year or so before Angular 2 is in a release-ready state... It's a huge break from the prior version, and imho a good one. The tooling will be the most interesting to me, I can only guess that they'll use node and traceur for building the output in a project to something a browser can use. I've been moving towards React (and Flux) as it lines up well with the way I've been developing with node. This just takes it a step farther. As for the features that aren't well known, this will change and have to be learned. There's enough in the ES6 stuff that I'm running new development against Node 0.11 so that I can get generator support and some other language features in place. Though still using the CommonJS/Node style require() statements. I do hope, that like React and its' tooling that the Angular guys have a compelling case and support for shared server/client logic and rendering as well as support for Browserify and Webpack+hot-reload. Though I like Webpack I think it goes too far from the node system, it is really nice however and I can see why others use it. It's all really interesting... I'm curious to see where Polymer goes from here as well...

cageface 12y ago

Agreed. This looks almost like a complete rewrite, which seems like a good thing because IMO Angular 1 is overly complex. But it does make me wonder if I should bother to really learn Angular 1 at all. The main thing pushing me towards Angular has been the size of the community but if they're wiping the slate then I might just go with React after all.

tracker1 12y ago
Angular has some huge support and momentum behind it... there has also been a lot of job postings around it as well. That said, I couldn't get past some of my issues with it, and that server rendering (shared code) was not really a good option there... The issues with dependency injection is another issue, and testing pretty much requires a browser. These are the reasons I'm using flux/react now. I'm really looking forward to following the development of Angular 2 now, and may just be my tool of choice for new development in a couple years.

nawitus 12y ago

Google is developing AtScript, which compiles down to ES5. I don't think that's the right way to go, though..

dragonwriter 12y ago

> Angular 2.0 relies heavily on ES6 features and syntax which are not yet supported in any browser Which is why part of the project is a compiler which compiles ES6 (actually, an ES6 superset) to ES5, which is supported by every browser. https://github.com/google/traceur-compiler https://github.com/google/traceur-compiler