Not sure I agree with this video. Eric Bidelman says that Emberjs is heading the wrong way.
What is your feedback
He lost me at, “It doesn’t matter what spec it’s in.” Standards are what makes all of this possible. They matter quite a bit.
Very interesting video. More in-depth technical look at Web Components than I had seen before. I didn’t hear him mention that Emeber js is heading the wrong way. Where in the video?
He did however highlight how AngularJS is very much “in-line” with the Web Component approach, but I do think that Ember will integrate nicely with Web Components as well. I watched a recent presentation by @wycats where he talked specifically about how Ember was geared towards the coming HTML 5 web technologies, including native observers and web components. Surely, handlebars will likely be replaced by Web components, which is why it is nice to see, that Ember doesn’t feature a complete widget set which will only become obsolete in a couple of years and take two much focus away from the core engine parts.
1 Like
@thomasboyt, Yehuda has given this talk a few times, but I think this is the one @kmandrup is talking about: Building Web Applications with Ember.js - YouTube. He starts talking about Web Components somewhere around the 40 minute mark I think.
Having looked at AngularJS a few hours today, I must say that it is way more mature, has a much bigger community, more stable, better testing/debugging support and is more geared towards Web Components… it is actually hard for me to see what Ember JS offers that Angular does not, except for some magic creation of default routes, controllers and views if you follow some naming conventions… I think Ember might be a serious contender in 6 months time if enough developers support its continued development and the code base has a major cleanup and restructuring to be more modular and OO, but it looks to me like a tough uphill battle… perhaps the architects should shift course and target it as the best SPA framework for Rails/Django and not try to make it a “general purpose” SPA framework? If they could win those communities over it would be a majo win I think.
I personally feel they should have opted for CoffeeScript, even if using javascript keeps dependencies down and makes it easier to get started… also QUnit feels too much like Test Unit IMO. Too “low level” for a framework trying to be “high level” IMO. My feelings at this time… Sorry if I seem too critical… it is easy to critique when you stand on the sidelines and not doing the hard work and fighting the battles yourself 