The script that creates the listing uses a 'big hammer' approach of testing for the existence of properties with names taken from a long list (8500+ items) of possible (likely, or known to exist at least one browser) property names. This approach has the advantage that it finds properties that are not normally enumerated by browsers and has the drawback that it is extremely slow. All properties for which (typeof object[property]) does not return "undefined" are listed.

It is in the nature of this type of testing that I need a real HTML page in order to list it's DOM. The DOM listing presented is the JavaScript DOM representation of http://www.litotes.demon.co.uk/dom_test.html ( page source). However, I cannot put an example of all possible HTML elements into that page (it is going to look fairly messy anyway). I am going to try to put examples of all of the common HTML elements into the page but I also would like to keep the DOM relatively simple. Otherwise the task of collating the resulting pages will become impractical.