Problem with AngularJS 2.0 transformation and why we even use a js framework

195 views
Skip to first unread message

Adam Chen Wei

unread,
Apr 8, 2015, 7:45:43 AM4/8/15
to ang...@googlegroups.com
Recently I read a lot about ECMA6 and AngularJS 2.0 that is going to come with it. It really a hard hit for me to know that a new born framework just about 2 years, now is going to be abandoned in likely a year or two. Its a whole framework rewrite that does not support any legacy features from AngularJS 1.x. I saw many positive talk about "embrace the change", however, how feasible it is that, if a release of the improvement in a  language, that a framework is not able to provide backward compatibility of some sort to its many loyal users and decide to ditch the way, just because... its hard to do? Yea, the talk on short pain but better performance. Imagine all the codes that need to be rewrite to accommodate the new framework! This act by Angular has bring my thought on "why I even need the framework?" Does it really "helped" me to code faster and be able to maintain code easily and scale? I feel at this point, or at the point in the future when 1.x is ditched, thats when all these, will collapse like the twin towers. It could be a glorious fall. So, my thought bring into component libraries instead of depend on a 100% buyin of a full scale framework. As it seems, framework has lost its place, even the best one at the time --- Angular. Framework re-write simply because language improvement without backward compatibility is literally, I feel, a betrayal of users (hey, Angular team owes me nothing for sure), but we `did put energy, thought, and dedication into believing the framework can work out and ready to help after we master it. However, how the heck (excuse my language), you are expecting one to master it, while there is not really much time to do it (well, unless you are with Angular since it birth, however, most likely, those people are the ones who made it)? I also understand web is dynamic ,and technology is improving. However, how could, ever, this transition will be painless or even, possible, in the first place? AngularJS 2.0, in my opinion, a really dangerous move that is make a huge gap between two ralm. Why can't the framework be the bridge for them, instead of just another glorious entrance to another place that everyone has to jump up and fight for the new fancy land again and ditch everything built on the land (excuse me if the metaphor does not make sense lol)?

Do anyone feel the same disappointment for js frameworks and please share your thought on it.

Nelo Mitranim

unread,
Apr 9, 2015, 2:00:18 AM4/9/15
to ang...@googlegroups.com
Nope, disagree on all points. There's nothing wrong with abandoning old technology to move forward. They took it as far as they could, and now the old architecture is holding further improvement back. Angular1 isn't going away, all your web applications are going to continue working. You don't have to update, but you can choose Angular2 for some of your applications to benefit from it.

I'm horrified at your suggestion that we should always keep using old technology for the sake of backwards compatibility. With this mentality, the rest of the software world will leave you in the dust pretty quickly.

Adam Chen Wei

unread,
Apr 9, 2015, 9:11:35 AM4/9/15
to ang...@googlegroups.com
Ok, I think what I mean is that, why not AngularJS Team offer a transition strategy that able to allow AngularJS 2 be some sort of extension of AngularJS 1 for a while so people can migrate easily, deprecate certain stuff after a while. I feel the transition is a bit radical, that does not allow a building block effect. That's what I am trying to convey, and the fact that a framework rewrite is so easily, yet, applying a framework cost WAY too much since its affecting the whole system. Not if we impliment each function as some sort of libraries so that each of them can be detached from each other, while portion of them allow to fall but not all will fall AT THE SAME TIME, which AngularJS 2 does. It will discontinue all the effort that people building beautiful integration with Angularjs and start from group zero again. Once again, I just feel its not really thoughtful and rushy to  just revamp. Its easy to revamp anything, but I think its better, when there are many people use it, allow building block effect to achieve more.

Don't you think?

On Thu, Apr 9, 2015 at 2:00 AM, Nelo Mitranim <m...@mitranim.com> wrote:
Nope, disagree on all points. There's nothing wrong with abandoning old technology to move forward. They took it as far as they could, and now the old architecture is holding further improvement back. Angular1 isn't going away, all your web applications are going to continue working. You don't have to update, but you can choose Angular2 for some of your applications to benefit from it.

I'm horrified at your suggestion that we should always keep using old technology for the sake of backwards compatibility. With this mentality, the rest of the software world will leave you in the dust pretty quickly.

--
You received this message because you are subscribed to a topic in the Google Groups "AngularJS" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/angular/QkdROwzM3NQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to angular+u...@googlegroups.com.
To post to this group, send email to ang...@googlegroups.com.
Visit this group at http://groups.google.com/group/angular.
For more options, visit https://groups.google.com/d/optout.



--

Best and Peace,
Adam

Sander Elias

unread,
Apr 9, 2015, 9:30:22 AM4/9/15
to ang...@googlegroups.com
Hi Adam,

How did you get this idea? The team is still very busy fleshing out Angular2, but they already stated quite a few things you can do to prepare your code for migration. Migration won't be free, but at the same time it is not as tough as you make it to be. 
Lets start with the first step in any migration plan. The new router that is near being ready will enable you to load both NG1 + NG2 in the same app/page/spa. It will be able to route NG1 and NG2 parts seamless next to each other. This means, that you can do a part by part migration. You don't need to do a full migration at once. You can even make the choice to build your new Modules in ng2, and leave the rest as is. 
When you get the hang of migration you will notice that a lot of thing are much easier to do with NG2. And NG2 will enable you to use all kind of 3rth party web-components as if they where your own.
If you want to keep everything you build in ES5 (or even ES3) that is fine. You don't have to use ES6, you can harness the full power of NG2 with the JS you already know.

Hope this eases your mind a bit,
with kind regards
Sander

Adam Chen Wei

unread,
Apr 9, 2015, 10:31:20 AM4/9/15
to ang...@googlegroups.com
Ah Cool! That's what I really want to know. It will be so much easier for me to know that it actually can be running side by side, or even possibility of it. Great! Thank you very much for the very useful info!

--
You received this message because you are subscribed to a topic in the Google Groups "AngularJS" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/angular/QkdROwzM3NQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to angular+u...@googlegroups.com.
To post to this group, send email to ang...@googlegroups.com.
Visit this group at http://groups.google.com/group/angular.
For more options, visit https://groups.google.com/d/optout.

James Cook

unread,
Apr 9, 2015, 9:31:32 PM4/9/15
to ang...@googlegroups.com
A common misconception is that AngularJS is a new web framework. In fact, it is one of the oldest SPA frameworks in use today. Angular 2 will take advantage of many new browser advancements that never existed a few years ago, and some which are still polyfills today.

According to Wikipedia:
  • GWT (2006)
  • Angular (2009)
  • Backbone (2010)
  • Knockout (2010)
  • Ember (2011)
  • Meteor (2012)
  • Durandal (2012)
  • React (2013)
First tag in GitHub - 10/20/2010
Reply all
Reply to author
Forward
0 new messages