W3C home > Mailing lists > Public > www-html@w3.org > August 2002

Re: Comments on XHTML 2.0 Working Draft

From: Chris Mannall <chris.mannall@hecubagames.com>
Date: Wed, 21 Aug 2002 12:34:12 +0100
Message-ID: <3D637AB4.4020000@hecubagames.com>
To: www-html@w3.org

On the subject of the modified content model of the p element, I wrote:
 > Your reasoning about not indenting paragraphs after a contained
 > element is my reasoning behind saying I was fine with the idea of
 > lists inside paragraphs, but I'm still not sure about tables inside
 > paragraphs. I suppose it comes down to your established ideas of what
 > "paragraphs" can contain - and I would never be happy with the idea of
 > a table inside a paragraph.

Philip Taylor replied:
 > Surely it all depends how the author writes his/her prose; I believe
 > that an author /could/, with some justification, interrupt a sentence
 > to present a table, and then continue with the sentence once the
 > table is complete (I can probably find good scientific texts where
 > this has been done, if you are not convinced by arguments that
 > whatever is possible should also be permissible...). If the sentence
 > can be interrupted, then the para. containing that sentence /must/
 > be interrupted.

After some thought, I've come around to this way of thinking on this 
matter now.

Several things have caused my shift in view. I've seen examples of 
interrupted paragraphs, and have also had a number of discussions with 
people who have seen tables in paragraphs or would like to be able to 
use tables inside paragraphs, and so on. Although I still think I would 
be pretty uncomfortable with using such structures myself (more for 
stylistic reasons), I can see why people would want to be able to do it.

If this weren't enough (which it was), I've also realised that I do 
agree with the concept that a table can "belong" to a paragraph. For 

   <p>Blah blah lengthy theory blah blah. This is shown by the following
           <!-- snip table contents -->

This seems natural to me, and has end-user value as well in that e.g. 
software that extracted the first three paragraphs of a document for 
summary purposes would "know" to extract the table, and so on. Note that 
this isn't really the same thing as "interrupting" a paragraph, more 
like implying a relationship between a paragraph and a table.

I believe Dorethea may have hit the nail on the head when she said my 
view on this betrayed my lack of a print-publishing background ;-) [1]

- Chris.

[1] http://www.textartisan.com/caveatlector/archive/2002_08.html
     (Search for "XHTML" to locate relevant section)
Received on Wednesday, 21 August 2002 07:43:44 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:00 UTC