xhtml <body> w/ text() nodes-- why?

[Gordon - Adelphia]

I have a question regarding xhtml. Why, why, why does the ELEMENT <body> allow “unblocked” text. HTML does not (though, most browsers will render). Xhtml (transitional) however allows text nodes (PCDATA).

All of HTML 4, xhtml – transitional, and xhtml-strict allow <div> to contain “unblocked” text. Does anybody know why – I’m looking for the philosophy behind allowing such.

I’m in the process of encapsulating xhtml mark-up in content marked xml. The content mark-up describes technical “facts” – the descriptions are encapsulations of “ideas” expressed in English. Those encapsulations use most of the exposition tools available to authors: tables, images, … most all things allowed in xhtml. Now, for a variety of reasons I do not want text fragments just hanging – they need to be part of a paragraph (<p>), a list, a table… Until I started on this project I never paid close attention to the formal dtds defining xhtml/html variants – I just always thought “unblocked text” was not allowed. If it were not, I would be able to simply adopt large portions of those dtds in my dtd (and easily convert existing html and portions thereof into my compendiums of xml “wrappered” facts.

I’m hoping that if I could understand the reasoning rational for allowing “unblocked” text in a div (or, on the extreme case in body), I might see that I should not be worried about the same in my application.

Thanks for any insight you can supply.
Gordon

 

[Martin Honnen]

I have a question regarding xhtml. Why, why, why does the ELEMENT <body> allow “unblocked” text. HTML does not (though, most browsers will render). Xhtml (transitional) however allows text nodes (PCDATA).

All of HTML 4, xhtml – transitional, and xhtml-strict allow <div> to contain “unblocked” text. Does anybody know why – I’m looking for the philosophy behind allowing such.

Looking at the HTML 4.01 transitional DTD at
http://www.w3.org/TR/html4/sgml/loosedtd.html the BODY element is defined as <!ELEMENT BODY O O (%flow* +(INS|DEL) -- document body --> where flow is defined as <!ENTITY % flow "%block; | %inline;"> and inline is defined as <!ENTITY % inline "#PCDATA | %fontstyle; | %phrase; | %special; | %formctrl;"> thus I think that with HTML 4.01 transitional it is valid to have text nodes as child nodes of the BODY element. An example is at http://home.arcor.de/martin.honnen/javascript/200406/test2004060501.html it validates with http://validator.w3.org:
http://validator.w3.org/check?uri=h...+automatically)&doctype=Inline&ss=1&verbose=1

Looking at the HTML 4.01 strict DTD at
http://www.w3.org/TR/html4/sgml/dtd.html
the BODY element is defined as
<!ELEMENT BODY O O (%block;|SCRIPT)+ +(INS|DEL) -- document body --> thus with HTML 4.01 strict text nodes are not allowed as direct child nodes of the BODY element.

Now looking at the definitions in XHTML 1.0 transitional at http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd the body element is defined as <!ELEMENT body %Flow;> and Flow is defined as <!ENTITY % Flow "(#PCDATA | %block; | form | %inline; | %misc*"> while in XHTML 1.0 strict the body element is defined as <!ELEMENT body %Block;>

I don't see where XHTML 1.0 transitional differs from HTML 4.01 transitional in regards to text nodes as direct child nodes of the body element, both allow. Furthermore both XHTML 1.0 strict as well as HTML
4.01 strict both disallow text nodes as direct child nodes of the body element. Thus I don't see the problem with body you have outlined above.