W3C home > Mailing lists > Public > public-html@w3.org > September 2009

Re: dt/dd in figure/details has killer rendering issues in ie6 and ie7

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Sat, 19 Sep 2009 03:02:29 +0200
Message-ID: <4AB42DA5.8020508@xn--mlform-iua.no>
To: "Tab Atkins Jr." <jackalmage@gmail.com>
CC: public-html@w3.org, Shelley Powers <shelleyp@burningbird.net>
Tab Atkins Jr. On 09-09-18 22.25:
[...  bike shed comment ignored ...]

> Curiously, the DOM of the page is just fine (as exposed by IE8's
> Developer Tools).  So it's just something in the styling subsystem
> that is crafting a malformed view of the document.

Live DOM Viewer shows wrong nesting in IE6/IE7/IE8-as-IE7. Hence I 
tested to see if CSS and DOM roughly corresponded, and the problem 
is that they did ... May be hasLayout is involved, and reason for 
the CSS irregularities.

> However, this error is a completely killer of the
> dt/dd-within-figure/details proposal.

If so, it would've killed <dialog> too. Who's done the testing ...


> It appears, then, that if we want <figure> and <details> to be usable
> on any reasonable timeframe, the only solution is a new element, or an
> attribute as I suggested elsewhere.

Or <dl figure><dl details>, as another "I" suggested ...

>  Pretty much all existing elements  have been exhausted at this point, I believe.

Is it acceptable to calculate with the need for wrapping? Fact is, 
if you wrap the <figure> inside another element, like in a <div> 
or in an <object>, then things works:


In IE6 and 7, the OBJECT element disappears in the DOM ... Same 
thing happens if you wrap the FIGURE element around OBJECT (and 
the DD/DT inside OBJECT again). So I guess using OBJECT as a 
wrapper could be another HTML 5 shiv ...

Wrapping is also useful for getting Firefox 2 support, though the 
OBJECT doesn't vanish in the air, there.

However, if one has to do all these workarounds to get reused 
legacy code to work, then it kind of confirms that reusing isn't 
the best choice.

Shelley Powers On 09-09-18 22.39:
 > Oh, well thank goodness there's now a _real_ reason why dt/dd 
can't be used.

leif halvard silli
Received on Saturday, 19 September 2009 01:03:12 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:51 UTC