WELL TOO LATE NOW HERE I AM
Originally Posted by Paul
This whole big stupid argument about what to put in <nav> is totally irrelevant. <nav> means nothing in browsers. It means NOTHING and browsers expect NOTHING inside them.
There is only ONE group who possibly gets ANYTHING from <nav> tags today. Who? Screen reader users. Nobody else. Nobody at all. And this is only because AT is reading the native role the <nav> element has. Well, we don't need <nav>s to get roles, and the WHATWG and W3C haven't even decided all the roles yet anyway. DRAFT etc yadda yadda
So, only screen reader users POSSIBLY get anything useful out of these new <nav> tags. And yet some folks here advocate filling them with a row of structureless links, or links with some random creative structure??? Why would you ever do such a thing? WHY? For some imaginary semantic purity? Purity for purity's sake? At the expense of users?? And no I don't care if you're addin in cute pipes or commas or what in between; everyone seems to understand this
is foobarz so luckily I don't have to rant about why this is retarded to the point of making sloths blush.
<a href="foo">Home</a><a href="bar">About</a><a href="baz">Blawg</a>
Making things more difficult for a group of people who already have a tougher time than the rest of us using the web and possibly NEED the web more than the rest of us aside, how about we AVOID the entire "problem" (fake invented problem) of <nav> tags entirely and just use the exact same thing <nav> does today in a way that has NOTHING to do with HTML semantics and other garbage argued about by web developers
There you go. This works. It works in all modern user agents. It works in most screen readers. Where it doesn't work, it STILL WORKS, because the users we are targeting are PEOPLE and they have experiences on the web and what have they experienced??? What do they know? What have they seen? What do they use? What is on practically EVERY. SINGLE. WEB SITE. OUT THERE. ? Lists of links.
Almost every web site out there uses a list of links to contain site navigation. To contain product navigation. People are used to this. People know how they work. Experienced screen reader users waiting for a page to load of a web site where they know they just want to get to the navigation, what is the first thing they do? Wait to see how creative you were with your <nav> tags???
They ask for a list of links on the page. They pull them up. They find the one starting with, usually, HOME. Boom. They have the navigation, and it's not all read out as one garbled chunk and it has not only LINKS in side them but also LIST ITEMS.
You can navigate not only by lists, and not only by links, but also by list items. These are common enough that a separate quick key for running through List Items is available.
List obsession? I counter that with what users expect and how they use websites (and a photo of a LOLCAT, for this is the Internet, good sirs). Users are people who take what they've already encountered and use it to help decide how they will navigate a new web site. The current paradigm is lists. Maybe next year image maps will come back into fashion and we'll be told to switch to some HTML5 version of those or something, but that's not today and that's not what people know. Web standardistas decided over the course of several years that navigation is best as a list of links and if you want to change that you WILL have mystery meat of one sort or another. Though confusing people might be the whole point of some web site, who knows.
Everything above changes if/when large numbers of sites switch to some other set of tags. Fine, but as conscientious web developers you will make sure users have already accepted it as their navigation paradigm before you make any such switch. I argue that today lists are still the paradigm the average user expects, and users of AT rely on.
Hey web developer peoples, you can do one of two things here:
we can insist that current HTML semantics sucks balls (so it does, oh well, I wish snow were purple and where is my goddamned p0ny???) and decide to use "better" semantics to go with the shiny new mostly-harmless-but-generally-useless HTML5 tags that real users have not actually experienced and makes things suck big hairy donkey balls for them (unnecessarily),
we can say, leave this stuff until it DOES have meaning and semantics and whatnot for browsers and other software, and until then, continue using the HTML structures real human beings have been using for years and are using today and will continue to use for the forseeable future and keep an eye on the new stuff for when it starts mattering to real users. "Real users" here is not referring to hipster nerds running Aurora and webkit nightlies because their idea of a good time is bug testing and brogramming, yo.
You people are free to argue til the cows come home whether you use a list in a nav or just throw a bunch of links or a p or a div or other weird creative things in there but if any of you actually write code like this and launch it I'll have to get old Gonz to rally up the blinks and we'll come storming your houses with pitchforks and refreshable braille displays and force you 1 year web access using only text browsers or assistive software so you can experience the wonderful creativity of "HTML semantics" on a regular basis. The "OH GOD NO WE'RE JUST BEING OLD FARTS EATING HEART PILLS AND LIVING IN THE PAST AND BEING SURPASSED BY DRAFT SPECIFICATIONS AND HIPSTER CODE OH NOES" doesn't convince me, but some concrete evidence-based arguments coming from actual user testing showing people benefit from other HTML totally would.
Actually I don't need to be convinced of anything, I'm not a web developer and I don't build these things, I just like to jump in and make messes for fun and profitz, lulz
Again I'm focusing on this minority because currently they are the only ones involved as users when it comes to tags like <nav>. That, and web developers who do CTRL-U and peek at web site underwear to measure your hipness level like the code voyeurs they are.
Oh and by the way... it may have occurred to several that the above statements can easily be used to defend sh*tty code practices that are used practically everywhere. And you can. And some do. Lately I've started to think that using the right tag today has more to do with "how do users use and perceive this?" than what someone on the W3C board said would be nice and p0nies 10 years ago. Yeah, midlife crisis, I'm losing my nazism... first I dump XHTML, now this... I may have to quit Python