- From: Ian Graham <igraham@smaug.java.utoronto.ca>
- Date: Thu, 13 Apr 2000 10:43:52 -0400
- To: "Braden N. McDaniel" <braden@endoframe.com>
- cc: www-html@w3.org, www-style@w3.org
On Wed, 12 Apr 2000, Braden N. McDaniel wrote > > ......... > > > > Yes. But they don't. And they never will do so it's not really relevant/ > > It's absolutely relevant. The spec permits it. You can't handwave it away > saying, "That'll never happen." The spec allows it to happen. It could > happen. > > But if the idea of this occuring in a UA style sheet is just too > far-fetched for you, perhaps considering this in the user style sheet will > bring it down to earth? The effect is the same. > > > > > Granted, this is an edge case. But the bottom line is that your assertion > > > > that an "empty" pseudo-class is unnecessary hinges on unspecified > > > > behavior. > > > > Something is necessary if it is useful. It seems that the reason that > > this is wanted is for sanctimonious reasons of 'Thou shalt not produce > > tag soup'. > > The "empty" pseudo-class has been suggested because, if the clause in > question in the HTML 4 spec refers to document *display* (styling) rather > than *interpretation*, we have a case where CSS does not fulfill the > formatting requirements of HTML 4. > > > As such the pseudo-class is useless. No-one would use it. > > *If* this is a style issue, browser authors creating HTML browsers using > a CSS style system ought to be using something like this. > Thank you, Braden, for an excellent summary. Given that Mozilla [1] (apparently) and Tasman [2] (definitely) have to implement special CSS code extensions to deal with empty paragraphs, it would seem sensible to introduce a standards-based way of dealing with this. In particular, if this is a layout issue for HTML, it will also be relevant to XML layout. Ian -- [1] http://lists.w3.org/Archives/Public/www-html/2000Apr/0114.html [2] http://lists.w3.org/Archives/Public/www-html/2000Apr/0115.html
Received on Thursday, 13 April 2000 10:44:14 UTC