element.write is like document.write, but for DOM elements. The core is John Resig's pure JavaScript HTML parser but it's been refactored so that incomplete HTML fragments can be submitted but still generate parse events, similar to the way browsers parse HTML and generate the DOM.

elementWrite.toElement(document.getElementByid('id')).
	write('<p id=hi>Hello <').
	write('i>World!').
	close();

Result:

<div id="id">
	<p id="hi">Hello <i>World</i></p>
</div>

Not exactly, but if you don't see a use case for it in your web apps, I'm not surprised. elementWrite is a library for libraries.

So how is this different from the original HTML parser?

elementWrite.toElement(document.getElementByid('id')).
	write('<p id=hi>Hello <'). # 1
	write('i>World!'). # 2
	close(); #3

1. At this point the DOM looks like this: <div id="id"><p>Hello </p></div>. i.e., the p tag and text are already in the DOM.
2. This completes the i tag and adds World.
3. If you used innerHTML for both of these fragments, you might get something like: <p id="hi">Hello </p>&lt;i&gt;World!, since innerHTML assumes you're giving it a complete fragment.

So basically element.write gives you the power to pause the HTML->DOM process at any arbitrary point and manipulate the DOM elements that are available.

See the tests for more interesting examples.