Hacker News new | comments | show | ask | jobs | submit login

This is a critique of the example, more than the framework itself. The "smart-float" directive is not a native angular directive, but rather a directive made for the example.

As for the "form.length.$error.float", it probably could be a bit more concise, but is it really that bad? What would be a "more semantic" way?

The main strength of angular, is that you are able to point your DOM elements to environment variables in your JavaScript code -- eliminating the need for traversing the DOM to find a mount node. If you are opposed to this functionality and insisting on traversing the DOM from the JS side, I'd love to hear why.

You can find small things you dislike in any framework, but rarely are they valid criticisms of the general concept. Often, these details are there for a reason, and just as often, they are flaws waiting to be fixed by an open source developer.




I'll keep this high level; that's where I have issues with Angular.

Angular wants you to learn its vocabulary, and encourages users to expand it further. Like you point out, smart-float is a directive specific to the example. To me, it is a foreign object I don't recognize in html. In fact, directives, scopes?

The big one. Dependency Injection, factories, modules. In JS, this is a solution looking for a problem. There are simpler alternatives to DI. They have issues, but only in theory.

Clarity matters a lot. ng-show="form.length.$error.float". Is there a more simpler way of doing it? There should be; I can't even tell what that means.

I think there are distinctly two camps. I'm in the one which wants frameworks to make me productive with the languages, specs and standards I already know. React (+ router) does this. Backbone does this too, depending on what you're building.

(Edit: replaced semantic with clarity)


I'll agree it's not pretty to view source and see an inordinate amount of fluff in what was once your beautifully handcrafted semantic HTML. Much of this can be alleviated with templating (ng-view).

Have you taken the time to play around with angular before airing your grievances? My hunch is yes and that you may just be simplifying things for others who haven't used it yet.

For those of you who haven't had the pleasure of working with angular: Once you've familiarized yourself with the verbiage, most of jeswin's concerns here come second nature just by glancing in one's source code. Let's take the example:

    form.length.$error.float
The name form, albeit horribly chosen in this case, represents the parent form's identifier in angular (the form's name attribute). It would've made more sense to use <form name="formLengthValidationDemo"> or something more insightful so you could more easily see the name matching from HTML to JS.

$error is simply the representation of any form validation errors that may have occurred. It's standard in angular, and there's plenty of documentation. Just google for "angularjs form validation" and you'll find a million resources at your disposal.

form.length.$error.float is checking if an error exists for the validation rule float. Given the limited amount of code in the sample, we'd have to arbitrarily assume that the "smart-float" directive is implicitly setting a custom validation rule via:

    ngModel.$setValidity('float')
I'll have to agree to disagree regarding pattern usage and dependency injection. They do have a place, and it tends to be in larger applications which become a burden to maintain as your application grows. To be quite honest, I believe angular brought a simplicity to DI that has yet to be seen in many other frameworks. It entirely lowered the barrier to entry for writing testable code and heavily promoted unit testing at the same time.


Agree 100% with your post - some of these complaints are misrepresenting the aspects that are not as desired in Angular. I've used Angular quite extensively for the past 1 1/2 years, and most of the complaints I have with it are more nuanced, such as having to dynamically inject some custom directives via $compile in a service method provided to a controller due to the complexity of being able to alter the attributes present dynamically in a template.

The declarative style that Angular coaxes you into makes things incredibly easy to understand - you can write code like

    <header>
      <navbar></navbar>
    </header>
    <main>
      <sidebar></sidebar>
      <content></content>
    </main>
    <footer>
      <company-info></company-info>
      <credits></credits>
    </footer>
And instantly understand what the purpose of these elements are - they're easy to read, they're descriptive, and they're clearly custom elements not part of the normal HTML5 spec, so you can instantly recognize that they must be directives.

I can also do simple directives like <loading-spinner></loading-spinner> and create a nice template for CSS to hook into by using the loading-spinner element tag without having to distract from the rest of my DOM with a nesting of 8 divs that can easily be tucked away in a reusable template.


I agree with you that since your elements appear to not be a part of the HTML5 spec they must be custom and are directives, wouldn't it be a good idea to mark them as such in some way? Maybe prepend them with "an-" or something?

The reason being, considering that HTML5 is an ever evolving spec that it might be possible for a new element to be introduced that could be using the same name as one of your examples. That might muck things up a bit. At least visually, as I would imagine the directives would likely continue working the same as before. It's just that the "they must be directives" logic would fail in that instance.


Good point. I, too, enjoyed using custom tags like <zoomer> or such, but I'm going to think twice in future.

Interestingly, out of the box Angular provides directives as a, input, or form—meaning that when you write <input> you are, in fact, invoking a directive that provides extra features along with native tag's functionality. While using Angular, you may as well consider every HTML element a directive.

Nevertheless, Angular uses ng- prefix with their built-in directives that don't replicate some existing tag's functionality closely.


You can make your custom directives HTML5 compliant, see [1], you can use the HTML5 data- prefix. For instance:

  <sidebar></sidebar>
can instead be coded as:

  <span data-sidebar></span>
or something similar.

[1] -- http://docs.angularjs.org/guide/directive


Custom elements are part of the HTML5 spec too:

http://w3c.github.io/webcomponents/spec/custom/

The only requirement is that they have a dash in the name. So while <sidebar> is not a valid HTML5 element, <a-sidebar> would be.


That's my usual way, or the class-based way for older IE.


Maybe prepend them with "an-" or something?

Yes, this is actually the recommended practice from the documentation. Built-in Angular directives are typically prefixed with `ng-`, and they suggest that other projects/developers use their own prefixes to avoid conflict.


That seems like a poor reimplementation of XML namespaces.

Why not use XHTML at that point? If XHTML ever had a clear mission statement in life it was that.

I mean I thought XHTML was a curse a few years ago when it was The Thing Every Good Designer Uses™ since it couldn't be rendered as XML in the most popular browser. But hopefully that's not as much an issue anymore. Or maybe it is. I haven't kept up with IE.


>That seems like a poor reimplementation of XML namespaces

Or you know, an adoption of only the needed part, instead of all the BS junk that comes with XML namespaces.

>Why not use XHTML at that point?

Because of all the other junk that comes with it. And because it comes from W3C.


XHTML is on the way out, I'm afraid. Angular is designed for HTML5.


Actually, it's a little more complicated than this even. Not only might HTML add more tags, but a non-standard tag might be a W3C custom element: http://www.w3.org/TR/custom-elements/

When users gain the ability to define their own tags, you might not know which ones are driven by directives, the custom element, or both even.


The philosopher David Lewis had a lot of critics, and they would often cite the fact that they couldn't possibly see how modal realism could be true, to which Lewis would reply, "I don't know how to refute an incredulous stare."

Arguments from personal incredulity don't engage substantively with what they ostensibly claim to refute because they import a variable- the argument-maker's conception of what seems plausible- that really should do no work in a logical analysis.

The arguments adumbrated in the parent comment seem to have similar failure modes, adorned as they are with premises such as "to me, [the directive] is a foreign object I don't recognize," and "I can't even tell what that means." Someone who has never programmed before could look at some vanilla javascript and plead perplexity, but obviously this isn't an indictment of javascript itself. To really make the point, the parent comment would have to show that the learning curve is disproportionately arduous for the payoff at the summit, but reiterations of personal unfamiliarity were offered instead.

Jeswin gives no evidence of having made a reasonable attempt to understand Angular ("directives, scopes?"), but more crucially, displays little understanding of the benefits of Angular. DI is criticised as "a solution looking for a problem" in the context of javascript, as if dependency injection were a bandaid for not being a dynamic language. DI enables testability, and not even mentioning this advantage, let alone not providing a cogent argument for why the learning curve of DI doesn't justify the gains in testability, is to refuse to participate in the discussion entirely. Directives are similarly dismissed as "foreign objects," which they might be to some, but they're hugely powerful foreign objects that allow arbitrary behaviour to be written declaratively. Not knowing the benefits of Angular obviously contributes greatly to a reluctance to learn it, and perhaps a subsequent rationalisation of this reluctance as being provoked by something intrinsic to the framework.

Fittingly, after expressing bewilderment regarding Angular scoping, Jeswin goes on to divide the world into two camps, declaring membership in "the one which makes me productive with the languages, specs, and standards I already know", which is about the closest thing to a natural language scoping bug I've come across.


> DI enables testability, ..... the learning curve of DI doesn't justify the gains in testability

Dependency Injection is a redundant, useless pattern in dynamic languages; especially when it comes to testability.

In test.js:

  //paymentService may be an instance or a class
  services.paymentService = mockPaymentService;  
  ....
  //use paymentService in your tests here.
The above over-simplified example satisfies most test cases. You can convince me with some use-cases.

(Edit: You weren't misrepresenting what I said. Removed that claim. I misread, apology.)


With DI you can ensure you're not accessing anything that's globally defined. While you can mock out global objects and methods on global objects, it becomes far less clear what the ramifications of doing so becomes when your app becomes large and complicated. With DI, you have all the dependencies clearly defined for the function you're concerned about.

Beyond testing, DI makes it much easier to swap in different implementations of a dependency than simply throwing things into the global scope lets you do. Suppose I want to switch my model from one that communicates with Parse to one that communicates with Firebase. Supposing I adhere to the same interface, I should be able to simply drop it in place without affecting other functions or objects that still rely on the Parse version of the model.


Your first point cannot be debated, since we are speculating. In practice, I'vent seen a problem. It works, it's clean, and it is the right amount of engineering needed to solve the problem.

Parse and Firebase example:

  //someservice.js
  ...
  //works as long as both retain the same interface.
  provider = new FirebaseProvider; //new ParseProvider()

  //add: just using an if condition, the old fashioned way.
  getProvider: function() {
    if (config.provider === 'parse')
      return ParseProvider()
    else
      return FirebaseProvider()
  
Why is writing configuration cleaner than doing the above? That thought process is carried over from projects with lengthy build time.


Whether or not Angular's implementation of its provider is cleaner is irrelevant to your original point that DI is a "problem looking for a solution" in dynamic languages.


My interpertation of the AngularJS DI implementation is that it exists primarily to make it easy to swap items out for testing purposes. I believe it solves that problem very well.

What alternatives do you recommend to DI?


Nothing, for swapping things out. :)

In my test initialization, I'd probably just replace the value of "ns.someService" worldwide with a mock.


> just replace the value of "ns.someService" worldwide with a mock.

Having a window-wide namespace might make sense in the test, but it is pretty risky/limiting if you are writing a component that should be reused in contexts you don't fully control. There are other ways around that besides DI, but they will probably have similar trade-offs in safety EOU, etc.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: