new polymer framework

[Brandon Pedersen]

Yet another framework from Google that looks pretty interesting.

http://polymer-project.appspot.com/getting-started.html

Kind of like custom directives in Angular but looks much simpler :), not quite as full featured but they are just getting started.

-Brandon

[Jamund Ferguson]

Yeah it looks awesome. The <link> syntax for templates is so much nicer than the angular templateUrl thing. I wonder if this has decent module support or built-in optimizers or something. Either way web components are looking pretty good :)

--
Group meets on the 1st Thursday and 3rd Tuesday of each month. See http://utahjs.com/location/ for more info.
---
You received this message because you are subscribed to the Google Groups "Utah JavaScript" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utahjs+un...@googlegroups.com.
Visit this group at http://groups.google.com/group/utahjs?hl=en-US.
 
 

[Dave Smith]

Are you talking about this?

    <link rel="import" href="toolkit-ui/elements/g-menu-item.html">

I think that is orthogonal to Angular's templateUrl, if I'm understanding you correctly.

--Dave

[Jamund Ferguson]

It's very likely I don't know what I'm talking about, but I like the look of the Polymer syntax :)

[Ryan Florence]

Polymer just stole angular's marketing angle of polyfilling the web.

https://github.com/Polymer/docs/commit/882691d4ba9c59aa15c4991dcd88eb0e0f9a23e4#L0L14

Looks like somebody from the angular team took offense. They word-smithed it, and eventually force-pushed over the top of this commit (but the sha can never die!)

Its noteworthy this is not an official google project, but done by "google employees". Perhaps google employees that don't like angular. Disclaimer: I don't work for google.

This troll brought to you by

- Ryan Florence

[Clint Berry]

Ryan, your theory about someone at Google not liking Angular is seriously flawed, since EVERYONE likes angular.

On Sun, May 19, 2013 at 12:21 AM, Clint Berry <cl...@system-11.com> wrote:

Ryan, your theory about someone at Google not liking Angular is seriously flawed, since EVERYONE likes angular.

--

[Bret Little]

Definitely looks interesting. Looks like someone could put together a polymer example on TodoMVC.

[justin]

The cool thing about a proposition like that is that it is so easily disproven:

I don't like angular.

I am a subset of "everyone".

Therefore, nope. QED.

--justin

On Sat, May 18, 2013 at 11:23 PM, Clint Berry <cl...@beingberry.com> wrote:

Ryan, your theory about someone at Google not liking Angular is seriously flawed, since EVERYONE likes angular.

On Sun, May 19, 2013 at 12:21 AM, Clint Berry <cl...@system-11.com> wrote:

Ryan, your theory about someone at Google not liking Angular is seriously flawed, since EVERYONE likes angular.

On Sat, May 18, 2013 at 11:26 PM, Ryan Florence <rpflo...@gmail.com> wrote:

Polymer just stole angular's marketing angle of polyfilling the web.

https://github.com/Polymer/docs/commit/882691d4ba9c59aa15c4991dcd88eb0e0f9a23e4#L0L14

Looks like somebody from the angular team took offense. They word-smithed it, and eventually force-pushed over the top of this commit (but the sha can never die!)

Its noteworthy this is not an official google project, but done by "google employees". Perhaps google employees that don't like angular. Disclaimer: I don't work for google.

This troll brought to you by

- Ryan Florence

--
Group meets on the 1st Thursday and 3rd Tuesday of each month. See http://utahjs.com/location/ for more info.
---
You received this message because you are subscribed to the Google Groups "Utah JavaScript" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utahjs+un...@googlegroups.com.
Visit this group at http://groups.google.com/group/utahjs?hl=en-US.
 
 

--
Group meets on the 1st Thursday and 3rd Tuesday of each month. See http://utahjs.com/location/ for more info.
---
You received this message because you are subscribed to the Google Groups "Utah JavaScript" group.
To unsubscribe from this group and stop receiving emails from it, send an email to utahjs+un...@googlegroups.com.
Visit this group at http://groups.google.com/group/utahjs?hl=en-US.
 
 

--
http://justinhileman.com

[da...@thesmithfam.org]

An alternative explanation is that you do not exist.

[Ryan Florence]

Does his email not prove he thinks, therefore he is?

[Dave Smith]

Assuming the universe is deterministic, yes. But that's not an assumption I am willing to make ... yet.

--Dave

On May 19, 2013, at 4:45 PM, Ryan Florence wrote:

> Does his email not prove he thinks, therefore he is?
>

[Paul English]

I like Angular, and I like Ember (I use Angular most). I really like web components, or at least what I've seen so far.

Just in case you didn't see these, both angular & ember have started to think about web components at some level.

https://groups.google.com/forum/#!msg/polymer-dev/4RSYaKmbtEk/uYnY3900wpIJ

https://gist.github.com/wycats/9144666b0c606d1838be

Thanks, http://www.2ality.com/2013/05/web-components-angular-ember.html

I'm betting we'll need less and less of angular and ember in the future.

[Ryan Florence]

Yeah, great link.

> both angular & ember have started to think about web components at some level.

Both have been thinking about web components since web components have been thought about.

As for ember, I've talked a lot to Yehuda/Tom about this a lot over the last few months and the semantics of ember templates/views map 1-to-1 to web components.

Come to lunch.js (formerly ember lunch) tomorrow at instructure and I'll be talking about this exact thing :) So I won't spoil it with code samples right now, but I'll post the code tomorrow.

[Seth House]

On Mon, May 27, 2013 at 10:30 AM, Ryan Florence <rpflo...@gmail.com> wrote:
> Come to lunch.js (formerly ember lunch) tomorrow

Ryan, do you have powers to update the lunch.js Meetup page for
tomorrow's meeting?

[Randall Bennett]

@ryan specifically, but everyone more generally:

Once web components are out, along with ECMA6 getter / setter watching, does the need for Angular, et. al., go away? Or just lessen?

[Dave Smith]

Having not reviewed the specs very closely, I think Angular's two-way binding will still be quite valuable.

--Dave

[Randall Bennett]

Dave, the getter setter stuff in ECMA6 (maybe delayed?) will absolve the need for dirty checking. Basically you can create listeners on individual vars, which JS will run when something is set. My understanding, didn't find the spec in the last 10 minutes.

[Dave Smith]

Right, so that will just make Angular faster at what it does best, which is bind expressions in markup (e.g., <div>{{foo}}</div>) to JavaScript variables (e.g., $scope.foo).

But it certainly won't replace Angular in my view.

That's topic #1.

Topic #2 is web components. These might fully replace Angular directives, which are a subset of Angular's overall functionality.

--Dave

[Seth House]

On May 27, 2013 12:47 PM, "Randall Bennett" <randall...@gmail.com> wrote:
> Once web components are out, along with ECMA6 getter / setter watching, does the need for Angular, et. al., go away? Or just lessen?

Lessen.

The end-goal for Polymer is to enable MDV (model driven views). The Google employees that introduced it at Google IO admitted in the q&a that Google is the only browser vendor pushing for MDV.

That style is very Angular-like (two-way data binding, plain JS objects, custom elements, DOM-based templating). It does not come with any direction on code organization, routing, or other helpers (promises, ajax, data persistence). So there's plenty of room left for a framework like Angular.

Angular did benchmark their stuff using Object.observe() instead of dirty checking and it was 40x faster.

Side note: Object.observe() will not make the ES6 spec. :-(

As Yahuda said in the Google Groups link in OP, the semantics (not syntax) of Ember match custom elements in web components more closely than Angular's directives. Both can be modified to use web components, of course. I'm interested in seeing Ryan's presentation tomorrow.

[Ryan Florence]

Once web components are out, along with ECMA6 getter / setter watching, does the need for Angular, et. al., go away? Or just lessen?

Like others have said, it just makes them smaller.

Ember's primary emphasis is providing an application framework, and secondarily a component framework (while angular feels more in the middle, or maybe closer to components since the emphasis is declarative HTML).

So when the component framework is native, something/somebody is still going to have to:

- deserialize a URL into an application state,

- transition from one state to another, 

- serialize the state to the url,

- manage view hierarchy, 

- manage data from an http service, and

- keep views up to date with data.

This is why I love Ember. The focus is the framework, more than the components. Components are the easy part, really. I've been building behavioral/declarative html for years, the hard part is managing application state, which Ember and Ember Data do so incredibly well.

[Seth House]

On May 27, 2013 1:00 PM, "Randall Bennett" <randall...@gmail.com> wrote:
> My understanding, didn't find the spec in the last 10 minutes.

My slides from Utah JS Conf are useless standalone. However my slides notes are packed full of links to specs, blog posts, and presentations, many notes, and further reading. If anyone is interested:

https://github.com/whiteinge/presentations/blob/master/utahjs_conf_2013-05-17_web-components/presentation.rst

[Clint Checketts]

On Mon, May 27, 2013 at 2:01 PM, Ryan Florence <rpflo...@gmail.com> wrote:

Once web components are out, along with ECMA6 getter / setter watching, does the need for Angular, et. al., go away? Or just lessen?

Misko's comments on the 'Breaking Open' interview (http://www.youtube.com/watch?v=X0VsStcCCM8) hit the nail on the head. Angular's opinionated approach will continue to be useful, even with things like Object.observe, etc.