Where do elements go?

As a response to Carst van der Molen, who commented on my latest article, I've written some tests to check where HTML elements go when you're leaving out HTML, HEAD and BODY tags.

I'm using the word "tags", because it are, in fact, only the tags that are left out of the document. The actual elements are inserted by the user agent.

Carst van der Molen made the following interesting statement:

Besides, I don't think you have any control what goes in the head and what goes in the body, which can be relevant for JavaScript execution if I'm not mistaken.

I've written some quick tests to see what exactly happens to elements when you leave out the popular containing tags.

The first one can be found here. The source is not very complicated, check it out to see what happens in detail. In summary; I've added some random elements to the document and a simple piece of Javascript loops through every element that exists inside the HEAD element, after which it prints the nodeName of that childnode to the document.

Test results for Firefox

Interestingly, it apparently just makes some educated guesses of what goes where. SCRIPT elements for example, go inside the HEAD by default. Even if the SCRIPT element resides at the very bottom of the document source!. The same is true for STYLE elements.

Test results for Safari

Safari gives back some odd results. The SCRIPT elements do not show up in the HEAD element! To make it sure they really aren't there, I've written another script. In Firefox, the word "SCRIPT" gets printed two times. In Safari, the page remains empty.

This was very unexpected to me, and I still don't really understand why this happens.

Test results for Internet Explorer

IE versions 5, 6 and 7 all give the same results as Firefox.

Test results for Opera

Opera does not give any results. It shows an empty document. This probably means it's a little more strict when reading the DOM than other browsers and does not make the left-out elements available to Javascript.

Final results

It looks like Carst had a fair point when questioning the control one's having over where elements go. The best thing to do is probably just to add the HTML, HEAD and BODY tags to your document when writing HTML. On the other hand; I can't think of much examples where I absolutely need to know whether elements exist in the HEAD or the BODY. 9 times out of 10 I reference elements either by tag-name or id.

In conclusion I want to point out that the W3C validator makes the same educated guesses Firefox makes. Validate the following example and you'll see the validator tells you it's an invalid document (due to a STYLE element being placed outside the HEAD element). If, however, the STYLE element comes before any other element that is considered to reside inside the BODY, it does actually validate.