Comments on: Semantic navigation with the nav element

By: HTML5: tag nuovi e anteprima delle innovazioni | paitadesignblog

HTML5: tag nuovi e anteprima delle innovazioni | paitadesignblog — Sun, 07 Mar 2010 11:05:39 +0000

[...] raggruppa gli elementi di navigazione [...]

By: Bruce Lawson

Bruce Lawson — Mon, 07 Dec 2009 08:24:38 +0000

In the example you give Gary, the Table Of Contents would be in a nav element (it is a TOC with links around your site). So an AT that’s sniffing “main content” would pass over the TOC, regardless of its position within the visual page or source.

The MSFT screenreader argument is entirely unrelated as far as I can see.

By: Gary Turner

Gary Turner — Mon, 07 Dec 2009 02:23:07 +0000

I confess a closer reading of your comment would have softened my own a bit. However, the page structure is still affected. Consider this top level sub-site page. If I had floated the toc, it would follow the “major” nav element and the header; not exactly the advantage you speak of. In order for your scenario to work, the page structure must still consider your, as yet, non-existent AT or PDA functionality. Hell, does MSFT’s screen reader make the intrinsic connection between labels and their nested form controls yet? That standard’s only been around for ten years.

Nevertheless, the nav element is still redundant, and brings nothing to the table not otherwise doable.

cheers,

gary

By: Bruce Lawson

Bruce Lawson — Sun, 06 Dec 2009 17:28:50 +0000

“Wow! And wouldn’t that cause an intolerable limitation on a document’s structure.”

Not sure I understand what you mean – what is the limitation, and for whom is it intolerable?

By: Gary Turner

Gary Turner — Sun, 06 Dec 2009 00:07:54 +0000

Bruce Lawson spoke at December 5th, 2009 at 11:53 am

> I disagree Gary.

> An assistive technology can make its own “skip links” function; it just needs to go to the first element on a page that is not header, footer, nav. That’s the main content.

Wow! And wouldn’t that cause an intolerable limitation on a document’s structure.

> It’s not just for assistive tech; I would love to see my phone do this for sites so that it scrolls to the start of the content for me.

> ATs and phones can then have a hotkey which goes to nav – by locating the first/ next nav element in the document as that is unambiguous.

For the most part, I dislike microformats as trying to do what xhtml’s author-described elements, or even html5’s new elements should do. This is a case where standardized class attributes would do the job as well as an element, without being redundant.

Good luck with getting ATs to give you the functionality you desire. Most can’t even deal with the implicit connection between labels and the form controls nested in them. That standard’s been around for ten years or so, and is much simpler to recognize than the usage you describe.

cheers,

gary

By: Bruce Lawson

Bruce Lawson — Sat, 05 Dec 2009 11:53:22 +0000

I disagree Gary.

An assistive technology can make its own “skip links” function; it just needs to go to the first element on a page that is not header, footer, nav. That’s the main content.

It’s not just for assistive tech; I would love to see my phone do this for sites so that it scrolls to the start of the content for me.

ATs and phones can then have a hotkey which goes to nav – by locating the first/ next nav element in the document as that is unambiguous.
,

By: Gary Turner

Gary Turner — Fri, 04 Dec 2009 09:49:43 +0000

@Alohci: By saying the nav element was equivalent to a div of class nav, I was saying that neither are worth a hill of beans. You bring up assistive technologies. Unless you depend on every possible client recognizing the nav element, and magically knowing when to skip that segment and move presciently to the right spot, you will still require a link that targets a particular id. If you have the link, and you have the target element, what earthly good is a nav element? It is plain and simply redundant, bringing nothing of value to the table.

cheers,

gary

By: Alohci

Alohci — Tue, 01 Dec 2009 00:48:20 +0000

@Gary Turner – Whenever the use of semantic elements is questioned, invariably one ends up discussing assistive technology (AT) and in particular screen readers. What

has that

does not is that because

has publicly defined semantics, AT can make use of that to easily allow the skipping of the entire navigation list and thereby go directly to the page content. For users who must experience or navigate the page serially, this is a real boon.

On the other hand,

does not have public defined semantics, it may mean “this is a navigation section” or it may imply something else entirely that means something sensible to the page author, in the author’s native language. For any given class name, AT is unlikely to guess unless the false positive rate is extremely low. So for the benefit of the widest possible range of users,

By: Gary Turner

Gary Turner — Mon, 30 Nov 2009 01:35:16 +0000

This proposed nav element is an example of paving the cow path before any cows have trod it. I have yet to see an example where the element wasn't superfluous, or didn't make the markup less semantic and less appropriately structured. When would navigation not be a list? By definition, tables of content, menus, bills of fare, &c. are all lists. How would an additional container that is the one-for-one equivalent of

improve the structure or meaning of the navigation list? John Faulds said, … I still want my ‘lists’ of navigation items to look like lists. I agree, and cannot see how a nav element will improve on the structure any more than a non-semantic div element would.

By: Jonas

Jonas — Sun, 18 Oct 2009 23:36:00 +0000

To me the nav tag is defined as loosely as it is because we are talking semantics. Semantics can be very subjective, and so it would make sense that the creator of a site should have the posibility to decide which of the navigation elements are major in his context.
As can be seen above, there are different interpretations and I think the comment made near the beginning – that seeing how users move on your side – might be a good firrst hint