stevenclark.com.au

Occasionally I see someone make perfectly valid pages which have failed to correctly use the heading element to create semantic meaning of the document. After all these are web pages and not web sites when we think of them technically - the protocols running are only concerned with the page requested and has no knowledge about the last page seen. That’s why we use cookies and other methods to remember.

H1 - H6 is not about the heading display size either. The look of the heading is nothing to do with which one you should use in your document. It is entirely about creating some semantic meaning within the information - check out the Semantic Extractor for an idea of your own document structure. Therefore the h1 itself should be about the meaning of the document itself - think of a tree heirarchy and you logically accept there can only be one title for the page (which makes sense). The discussion about whether that h1 is the site title, the page title or something else is beyond the scope of this article.

The reason I am writing this article is because in this grey area of common sense there will always be someone who has a divergent view of the semantics of something or who will pull out a bunch of gibberish and tell you (in a way you can’t argue back) that what they’re telling you is semantically correct. In truth I don’t think there is a document out there which defines the true semantic truth according to Moses. I’ve heard the arguments for two h1 elements on a page but to me it can never make sense. In fact to argue there can be - and I don’t care if its valid markup or not - makes me wonder if someone hasn’t just disappeared up the ring of their own rhetoric.

In fact I rather hate the word semantic as it has to be one of the most maligned and misguided terms used in our industry. Somewhat like the song Isn’t it Ironic not being about irony at all. Its just a word grown to be commonly used for a whole bunch of stuff not ironic at all - for example bad luck.

If by chance you look at your page and there is actually a good argument for two h1 elements you can logically conclude you should really be considering two separate pages for the information. If you have a logo and a h1 you decide should both be a h1 this usually means (to me at least) you are just trying to stuff your company name or site name before the other h1 as the first part of the real h1 and may even have more to do with you wanting your company or site name to be noted as more important by Google. If so then that’s another reason you shouldn’t be thinking of doing it - it should be about the meaning of the information not about grabbing the attention of bots to keywords. Either way I’m not entirely sold on the idea.

I accept my opinion on this might be different from some high profile standardistas (who I am notably not linking to for a hand to hand verbal stouche) and that is acceptable. We’ll just have to beg to differ. In the meantime just ask yourself if, regardless of anything else you hear, doing something really makes logical sense to the document itself.

This entry was posted on Wednesday, August 1st, 2007 at 3:46 am and is filed under web standards. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.