Even though the author didn't like it, I quite liked the ultra-chaining approach presented. So what if the readability relies too much on indenting, practically every language relies on indenting for readability (or the indenting actually has syntactical meaning like python).

Frankly, I found this example very neat and would love to see this in jquery.

  $('div') 
    .when('click') 
        .addClass('active') 
        .text('Hey') 
    .done() 
    // Also possible with on() 
    .on('mouseout') 
        .removeClass('active') 
        .text('Ho') 
    .done();

It relies on indenting for readability, just like everything else. I say go for it!

Though its a neat shortcut, a new member in the team who has not heard about the extension won't understand the code straight away. That's the problem of modifying the core behavior of a language/framework (often called monkey-patching).

That's an issue with every single piece of technology (operating system, library, application, API, programming language, CPU, non-trivial function or macro...). As such I don't think it can be a valid statement against the adoption of any specific technology. Besides, this particular concept is really easy to learn.

Of course, if all you want to do is add hover effects to your app, it's always worth considering doing it in straight CSS:

  .menu {
      background-color: #fff;
  }
  .menu :hover {
      background-color: #ccc;
  }

If this becomes the standard way of working with jQuery in a future release, I'll likely switch to MooTools or another library. Code like that would be hell to maintain.

It isn't being proposed as the standard way, it is merely a proposed alternative. Nothing wrong with other ways to do the same thing. Some prefer other forms of syntactic sugar.