I've just noticed that the long, convoluted Facebook URLs that we're used to now look like this:
http://www.facebook.com/example.profile#!/pages/Another-Page/123456789012345
As far as I can recall, earlier this year it was just a normal URL-fragment-like string (starting with #), without the exclamation mark. But now it's a shebang or hashbang (#!), which I've previously only seen in shell scripts and Perl scripts.
The new Twitter URLs now also feature the #! symbols. A Twitter profile URL, for example, now looks like this:
http://twitter.com/#!/BoltClock
Does #! now play some special role in URLs, like for a certain Ajax framework or something since the new Facebook and Twitter interfaces are now largely Ajaxified?
Would using this in my URLs benefit my Web application in any way?
This technique is now deprecated.
This used to tell Google how to index the page.
https://developers.google.com/webmasters/ajax-crawling/ (archive.org link)
This technique has mostly been supplanted by the ability to use the JavaScript History API that was introduced alongside HTML5. For a URL like www.example.com/ajax.html#!key=value, Google will check the URL www.example.com/ajax.html?_escaped_fragment_=key=value to fetch a non-AJAX version of the contents.