31 Semantic DIVs

I agree that aside, header, footer, etc. are useful, but do we need new elements for these? What about a role attribute? For example: <div role=“header”></div>. The role attribute (unlike class or id) could have a defined set of allowed values (aside, header, footer, etc.). This would add the semantics of the proposed elements while maintaining backwards compatibility with user agents that only understand div.

posted at 07:53 pm on December 4, 2007 by Liam Morland

32 HTML5 sounds nice but

what about the browsers catching up to the current specifications?
I agree with #27 on that. More focus should be on browser compatibility with the current markup of both HTML and CSS.

posted at 08:05 pm on December 4, 2007 by Misty Beier

33 HTML 5 *is* being implemented

A number of comments suggest that we’ll never see the benefits of HTML 5 but that’s simply not true.

1. <canvas> is implemented in Firefox, Opera (& Opera Mobile), and Safari (& Mobile Safari). It also has an extensive test suite that will help with interoperability issues. (Is it perfect? no. Is it in IE? no. bummer.)
2. <video> is implemented in the Opera 9.5 betas and in the WebKit nightlies.
3. WebKit nightlies implement HTML 5’s Offline storage API
4. The WebKit team has told the HTML WG that they want to implement the entirety of the HTML 5 spec in the very near future. (I wish I could give you a citation URL for this but I can’t find the transcript of the face-to-face meeting in Boston where Maciej Stachowiak talked about the WebKit team’s commitment to HTML 5.)

1. There is an open-source HTML 5 parsing engine written in Python & Ruby

There are more success stories if you follow the working group discussions (blogs, mailing list, IRC) and there is clear support from Apple, Google, Mozilla, and Opera to implement HTML 5 now, not later. Microsoft hasn’t quite joined the implementation game yet but is involved (Chris Wilson is the chair of the W3C’s HTML Working Group which is working on HTML 5 along with WHATWG.)

Is HTML 5 the solution to all of our HTML problems? Of course not, but HTML 5 is in a better position than where we were four years ago with XHTML 2 .

Finally I’d like to add that your opinions matter and will be seen by the W3C HTML Working Group & WHATWG. The spec editors (Ian Hickson and Dave Hyatt) are committed to taking not only feedback from the working group discussions but also feedback from the wider web such as blogs and publications like A List Apart .

It is really great that this discussion is happening in public across the internet and if you have ideas on how to improve HTML 5 then I hope you’ll stop by. On top of the links Lachlan provided in his article, there are some other public resources:

• relevant IRC channels

• HTML 5 issue tracker

Finally, we can always use volunteers who want to provide clearly defined use cases, test cases that prod the spec in every which direction, and people to help translate our treatise on browser behavior to web author accessible prose.

posted at 10:21 pm on December 4, 2007 by Shawn Medero

34 lowest common denominator (revisited)

Look, I’m all for advancing the Web. But we’re currently held back by IE on today’s approved standards. I would love to use some things available to me now – but I usually can’t because I have to support the 70+% user out there on it. Until that turd is flushed (as in dead) or polished (as in catching up with the rest of the kids), we’re stuck with the stink of old ways.

posted at 10:51 pm on December 4, 2007 by John Lascurettes

36 What forest? All I see are trees.

I have to say, I’m rather disappointed. Not at the amount of work which has gone into this (which is commendable), but rather the short-sightedness of those involved.

Listen to people like Bert Bos & you’ll hear them say that 13 years ago (!), they couldn’t even dream of how CSS would be used by people. They can’t even believe that it still exists. If CSS had been built to solve the problems of 1994 and only those problems, he probably would have been right.

13 years later, the HTML working groups involved in HTML5 have decided to approach things differently. The introduction of new elements like “header”, “footer”, “aside” and “section” reek of this, let alone the “article” element.

Extolling the virtues of “semantic markup” (as a general concept du jour) doesn’t cut it as an argument for these new elements. What real world benefits can we derive from their meaning?

More critically, what exactly is the goal of HTML5. I’ve heard some people say that the goal is:

more interoperability when recovering from errors [1]

That’s fair enough, but mixed in with those frothing at the mouth over these new elements, things are becoming more and more hazy as to what HTML5 is trying to solve. I’d like to hear a clear statement on that.

¹ Karl Dubost in Why HTML 5 Specification Matters?

posted at 12:57 am on December 5, 2007 by Nathan de Vries

37 So, when will HTML 4.01 arrive?

I am all for HTML V. It will give us Web coders and, of course, the browser makers, more tags to misinterpret and misapply. For sure, having the aside tag will allow me to semantical-up my pages where I have no possibility for that now.

I hope the HTML V spec takes a long time to put together because it will keep a certain class of person off the streets, safe from wandering dreamily, in mortal danger from passing vehicles.

Meanwhile, I am still waiting for a leading or non-leading browser to implement HTML 4.01. Fixed table headers / footers? Column alignment on language-specific decimals? Object? Cite? Char? And so on.

Not that this matters much. HTML is simply structure and our job is to keep structure as simple as possible. CSS, JavaScript, and other technologies build best upon simplicity. Usability, an ultimate goal, builds upon convention and repetition. Anyone complaining that an old spec or a particular browser is holding up progress should really be careful of passing vehicles.

posted at 02:51 am on December 5, 2007 by Brett Merkey

38 Clarification re WebKit commitment to HTML5

Sean Medero writes:

The WebKit team has told the HTML WG that they want to implement the entirety of the HTML 5 spec in the very near future. (I wish I could give you a citation URL for this but I can’t find the transcript of the face-to-face meeting in Boston where Maciej Stachowiak talked about the WebKit team’s commitment to HTML 5.)

Just to be clear, I didn’t make specific date or scope commitments, as Apple does not comment about future product releases. But I can say that we’ve implemented many aspects of the spec in WebKit nightlies, and we are interested in many other parts of it. In any case I couldn’t make a statement one way or the other about 100% support until the spec is final.

posted at 02:52 am on December 5, 2007 by Maciej Stachowiak

39 HREF, SRC?

Here’s what I’d love to see in the new version of HTML:

1. The href= attribute available for use on any element. If I want to make an element linkable, why do I either have to use javascript or put use an anchor?

2. An easy way to tie the height/width of an element to another element – the way table rows and columns tie height and width to groups of table cells. Some thing like: div {height: #header;}, which would grab the height of the #header element.

3. An easy way to vertical center in block elements (DIVs, etc.). How come table cells can do this, and but DIVs require give and take with padding and height/width?

Also, like a lot of other people in this board, It feels to me that the new elements <header>, <footer>, etc. are abstractions, not concrete realities. What happens if I have a header and subheader? What happens if I have two navigation bars? Why do I have to put <ul> INSIDE the <nav> block?

It feels to me like we’ll have to define what sorts of elements these things are – in other words, the header acts like a div, or the nav acts like an unordered list – which seems redundant to me. I don’t know. Why confuse things with unneeded additional elements when there are other, more powerful things we could be addressing?

Finally, I have agree – I know the process is necessarily long and laborious, but damn – years and years will go by?

:(

posted at 03:11 am on December 5, 2007 by Evan Sharp

40 All I want is readable code and effective, beautif

I think that the code, discussion and techniques discussed here in ALA ultimately mean more and have more beneficial effect than this HTML5 thing ever will.

I do understand that there is a syntactic difference between

<header>

and

<div id=“header”>

but for the life of me I do not see what desirable behaviors I cannot do with the second. And since divs share some behaviors, I find the second more understandable. In my cranky pseudo-object terminology, it means that header as defined in css inherits from div. So what if one is a tag and the other is an attribute value.

Although I am a student of language structure, I require meaningful application to my work which, in this context, is making web sites.

I also am guessing that there is going to be some paradigm shift, sea change, aha discovery or mere ‘this changes everything’ event that makes much of the hard work of HTML5 for naught. As others have mentioned (3,10,15) years is a long time.

OTOH, construction of The Cathedral of St. John The Divine was started in 1892 and it is still not yet completeted, yet it functions fully. Maybe HTML5 will be the St. John’s of the web.

posted at 03:20 am on December 5, 2007 by Brandon Sussman

Got something to say?

Discuss this article. We reserve the right to delete flames, trolls, and wood nymphs.

Create a new account or sign in below if you’d like to leave a comment.

Forgot your password?

Subscribe to this article's comments: RSS (what’s this?)