- From: Nir Dagan <nir@nirdagan.com>
- Date: Sat, 02 Jan 1999 17:34:47
- To: Ghadi Shayban <shaybang@yahoo.com>, www-html@w3.org
At 17:37 1/01/99 -0800, Ghadi Shayban wrote: > >Correct me if I'm wrong: > >In HTML 4.0, the body element couldn't have character data directly >inside of it. For example, this is illegal: ><html> ><body> >asdf ></body> ></html> It is invalid in the strict DTD but valid with the loose DTD. (well, the example is invalid as there is no TITLE before BODY) The differences between the content models of the strict and loose DTDs are not emphasized in the verbal part of the HTML4.0 spec. > >In WD-html-in-xml, these pieces of the DTD define the body element: ><!ENTITY % Block "(#PCDATA | %block; | %misc;)*"> ><!ENTITY % body.content "%Block;"> ><!ELEMENT body %body.content;> > >That states that the body element can have cdata directly inside. >This seems like a significant change from HTML 4. Can someone >enlighten me on this? In principle it is possible to write modular DTDs with different content models for strict and loose, but this requires overriding many entity declarations, and make it more difficult to follow. The strictness of the strict HTML4.0 dtd is more of "suggesting good practice" to authors as even with the strict DTD one can have a DIV as the only child of BODY and DIV allows for #PCDATA as well as %block;. Regards, Nir Dagan Assistant Professor of Economics Universidad Pompeu Fabra Ramon Trias Fargas 25 08005 Barcelona Spain http://www.nirdagan.com mailto:nir@nirdagan.com "There is nothing quite so practical as a good theory." -- A. Einstein
Received on Saturday, 2 January 1999 11:32:45 UTC