At ng-conf in the opening remarks they talked about how they plan to have people migrate from 1.X to 2.0. From what I understand if you start using the new router in Angular 1.4 then that will be the router for 2.0 which will allow you to use Angular 2.0 with 1.X components as you slowly migrate each 1.X component to 2.0. I think this is pretty reasonable.
> Not sure how they can "recommend" 1.X when it's completely incompatible with 2.0
I'm tired of people saying this. It's simply not true. The core team has said multiple times that there will be a migration path and they just announced today that you will be able to mix v1 and v2 components within the same project.
In addition to allowing your app to share components, they are also back porting parts of v2 to v1. So far, the router and ng-animate and I'm sure more will come.
I'm also guessing that someone is going to make an "add controllers to 2.0" addon that will be hugely popular with people slowly upgrading their 1.x apps
I think the idiom is "Cue the haters" but "queue the haters" does have a nice image to it. I'm imagining a long line of haters waiting patiently to dis Angular.
Piggy-backing off of this, here's my short list of if-js-framework-were-languages:
angular = php : obtuse and a little nonsensical at times, yet by far the most popular tool out there in industry.
ember = ruby : highly decorative and alluring, and it's even built by refugees of rails
react = haskell : esoteric, functional, and strange enough to come with its own dialect of js; it's literally the framework built for framework developers (to learn from)
backbone = C++ : fast because it has so little and therefore so little to slow it down
jquery = C : at some level, everything else seems to be built on this
meteor = computer language when portrayed on a Hollywood movie : you know, like in the first Wolverine Xmen movie, whatever language that they used to program Ryan Reynolds with that allows the user to just type in something like "Kill him" into the terminal and Ryan Reynolds then attempts to kill wolverine (sorry spoilers). I don't know if there are any commands other than "kill him", but it handles the "kill him" command with incredible automagic.
jQuery = C seems a bit odd. AFAIK none of those other frameworks are built off jQuery, though Angular supports working with it. Many are actually hard to integrate with jQuery.
They announced today that they're retiring the AtScript name and using TypeScript going forward. The new version of TypeScript is a combination of the old TypeScript and AtScript.
Interesting to see here that the docs still refer to AtScript, where there's an announcement from Microsoft on the homepage explaining that Angular2.0 will use Typescript.
White text on white background image ? Look how this renders on 1920x1080 resolution http://i.imgur.com/sPWpL6C.png. I cannot read that text, I don't understand anything from it.
Someone made a small mistake with linear-gradient. They have linear-gradient(top, #0143A3, #0273D4) instead of linear-gradient(to bottom, #0143A3, #0273D4).
Interesting... their language really seems geared towards pitching Angular as a rapid-prototyping tool. They explicitly mention UX designers, but not developers:
"Express your ideas with clean, understandable code. Angular is simple to build on, easy to change, and friendly to the way UX designers work. Create a UI that is beautiful by default, with material design and support for web components."
Shouldn't we be rapid-prototyping our ideas? Isn't that what "hackers" do often? "Hack" something together and then iterate like mad, failing fast on bad ideas and moving forward with the good ones?
I think a lot of us already see AngularJS as a development framework. A big part of developing something is also UX design. Do you have UX designers on your development team? I know that I do, and I can help sell Angular to them if they feel represented (as they should be).
Of course you should! I am a UX designer (sort of)! In fact, the only time I've used Angular was when I used the Ionic framework to create a proof-of-concept to test with users.
I was just commenting on the fact that they are marketing it as such. Previously, they didn't really seem to use such language.
I think it's a great move on their part, and it seems like it's a useful tool for rapid prototyping.
There are currently two posts on Angular on the front page of HN. Both have quickly devolved into multiple rants decrying "breaking changes", which can be traced back to the initial Angular 2.0 announcement: http://angularjs.blogspot.com/2014/10/ng-europe-angular-13-a...
This really speaks to the importance of communication around these announcements. If you pay attention to what's actually communicated, yes, there will be a migration plan for 1.x apps. Yes, 1.x will be continued to be supported for much longer than the lifetime of most of your little startups - sorry to break it to you. Yes, 2.x is a much-needed step forward for SPA app development. BUT, when Angular 2 was announced, it was announced too early - they shared the fact that breaking changes were coming, and that AT THAT TIME, there wasn't a plan in place for migrating 1.x apps. That poorly phrased communication has since dominated any discussion about Angular 2, very much to the detriment of the project. It's the same thing as the Flickr, Instagram, Facebook, etc TOU changes that caused such a brouhaha in recent years, except that you would think technically-minded developers would be able to parse the actual information.
I understand where you're coming from - but that singular announcement left a TERRIBLE taste in my mouth for Angular. I won't use it for any project, hobby or otherwise.
Compare it to Ember where they are bending over backwards to provide free improvements and a sane backwards compatible approach.
Why bet your entire business on a framework that has such poor communication skills? (Again: Assuming what you're saying is correct ;) )
Yeah, very valid point. Ember's outward communications are a really good model of how to deal with this stuff - unfortunately I just can't imagine myself working with the framework. The thing I find most hilarious is that the followup to the "OMG Angular 2.0 is making breaking changes" herdthink is usually "Let's all move to React 0.12!".
Wait, what? This is AngularJS? I'm a SysAdmin / DevOps guy that has recently taken up Web Development, spending the last year or so getting neck deep in AngularJS.
This looks and feels entirely different, and I don't care to start from scratch.
Angular 1.* will be around for awhile. It won't simply stop working. Eventually they'll stop adding new features, and eventually they'll stop fixing bugs. I don't know what the timeline is for that but it's far from immediate.
46 comments
[ 3.4 ms ] story [ 87.3 ms ] threadthere is a weird "rawgit traffic error" banner across the site.
Not sure how they can "recommend" 1.X when it's completely incompatible with 2.0. Might as well recommend Ember, React, or literally anything else.
I'm tired of people saying this. It's simply not true. The core team has said multiple times that there will be a migration path and they just announced today that you will be able to mix v1 and v2 components within the same project.
angular = php : obtuse and a little nonsensical at times, yet by far the most popular tool out there in industry.
ember = ruby : highly decorative and alluring, and it's even built by refugees of rails
react = haskell : esoteric, functional, and strange enough to come with its own dialect of js; it's literally the framework built for framework developers (to learn from)
backbone = C++ : fast because it has so little and therefore so little to slow it down
jquery = C : at some level, everything else seems to be built on this
meteor = computer language when portrayed on a Hollywood movie : you know, like in the first Wolverine Xmen movie, whatever language that they used to program Ryan Reynolds with that allows the user to just type in something like "Kill him" into the terminal and Ryan Reynolds then attempts to kill wolverine (sorry spoilers). I don't know if there are any commands other than "kill him", but it handles the "kill him" command with incredible automagic.
What am I missing here?
For reference: http://imgur.com/zdaMTFe
"Express your ideas with clean, understandable code. Angular is simple to build on, easy to change, and friendly to the way UX designers work. Create a UI that is beautiful by default, with material design and support for web components."
I think a lot of us already see AngularJS as a development framework. A big part of developing something is also UX design. Do you have UX designers on your development team? I know that I do, and I can help sell Angular to them if they feel represented (as they should be).
I was just commenting on the fact that they are marketing it as such. Previously, they didn't really seem to use such language.
I think it's a great move on their part, and it seems like it's a useful tool for rapid prototyping.
There are currently two posts on Angular on the front page of HN. Both have quickly devolved into multiple rants decrying "breaking changes", which can be traced back to the initial Angular 2.0 announcement: http://angularjs.blogspot.com/2014/10/ng-europe-angular-13-a...
This really speaks to the importance of communication around these announcements. If you pay attention to what's actually communicated, yes, there will be a migration plan for 1.x apps. Yes, 1.x will be continued to be supported for much longer than the lifetime of most of your little startups - sorry to break it to you. Yes, 2.x is a much-needed step forward for SPA app development. BUT, when Angular 2 was announced, it was announced too early - they shared the fact that breaking changes were coming, and that AT THAT TIME, there wasn't a plan in place for migrating 1.x apps. That poorly phrased communication has since dominated any discussion about Angular 2, very much to the detriment of the project. It's the same thing as the Flickr, Instagram, Facebook, etc TOU changes that caused such a brouhaha in recent years, except that you would think technically-minded developers would be able to parse the actual information.
Compare it to Ember where they are bending over backwards to provide free improvements and a sane backwards compatible approach.
Why bet your entire business on a framework that has such poor communication skills? (Again: Assuming what you're saying is correct ;) )
This looks and feels entirely different, and I don't care to start from scratch.
Angular 1.* will be around for awhile. It won't simply stop working. Eventually they'll stop adding new features, and eventually they'll stop fixing bugs. I don't know what the timeline is for that but it's far from immediate.