Use standard Grunt setup

[joefiorini]

I'd like to use ember-app-kit as our standard build tool for Ember apps. I'm happy to take on converting this over. Shouldn't change our usage at all, but will keep the NPM libraries and grunt plugins standard.

[joshwalsh]

So, it sounds like we need to convert this to use ember-app-kit and combine it with @bdielman baseline generator to be useful.

[joefiorini]

Yep, the baseline generator just generates the stylesheets, whereas ember-app-kit compiles whatever it sees in app/styles, so it shouldn't be a problem.

[bdielman]

Ember-app-kit is a great tool for setting up Ember apps, but is it overkill here when all we need is a prototyping tool? We'll want to modify and add our own stuff. How much of a pain will that be if we depend on ember-app-kit as the base?

Also my baseline generator does a little bit more than generate stylesheets. It sets up a basic Grunt workflow and starter template to create prototypes. 😃

[joshwalsh]

We don't need 90% of what Ember App Kit gives us, since I'm only using the Router and templates for prototypes. I'm using the auto generated controllers, models, views, etc.. so it stays lean.

That said, I still like Ember since it gives us the ability to do anything we want and not be limited by the constraints of many other prototyping tools. But, reaching into Ember more would be an exception more than a rule, until we all get more comfortable.

I'm not real comfortable with EmberAppKit, but I believe it's similar to what I have put together here.

[joefiorini]

The 400-line Gruntfile in this app is pretty complex already. I'm pretty sure we aren't using requirejs or Compass, but both are present, there maybe others. EAK does a lot of the same things, but does a very nice job of organizing task configuration. I don't want to do anything that would make anyone uncomfortable. I completely agree about reaching into Ember being a smell. I just think EAK provides a setup that will be easier to find our ways around.

I'll do a conversion in a branch and send a PR so you can see what I'm talking about (shouldn't take long). Then only if we're all comfortable with it will we decide to merge it in. If we agree it's too much, I'll be happy to kill it.