Angular uses dirty-checking, KO uses change listeners (which is why KO is objectively worse because change listeners cause a lot of problems)

http://stackoverflow.com/a/9693933/20003

Thanks!

I actually was looking more for articles that gave the philosophy behind Knockout though, not a comparison with Angular - I mentioned that more as a reference to my experience with client-side MVC. I would like to understand other frameworks independently from anything else - I want to know what they are trying to accomplish without any opinions.

I think it's fair to say that Knockout's philosophy is:

"Knockout is a library that binds Javascript objects to DOM nodes."

It accomplishes that task amazingly well. It essentially takes this DOM node:

    <a data-bind="attr: {href: link()}, text: title()"></a>

And let's you bind a JS object like this to it:

    var my_view_model = {
        link: ko.observable("#foobar"),
        title: ko.observable("A link")
    }

Then, knockout watches the `my_view_model.link` and `my_view_model.title` attributes for changes. When it "observes" a change of their value, it updates the DOM.

That's about all you need to grok to see value out of Knockout. If you can get your data into JS objects, Knockout can do a bang-up job of keeping your DOM in sync with your data. As a "backend guy" that eliminates a whole lot of boilerplate JS I used to write whenever my data changed or I needed to keep track of state.

Knockout does more, of course, but from then on it's mainly about adopting it to your use case (Do I mix it with Django templates? Do I just go with a static-only frontend? How do I manage AJAX requests to keep my JS objects data up-to-date?)