W3C home > Mailing lists > Public > www-html@w3.org > December 2003

Re: XHTML 2.0: Suggestion for <addr/> and <blockaddr/> to replace <address/>

From: Karl Dubost <karl@w3.org>
Date: Mon, 8 Dec 2003 10:39:42 -0500
Message-Id: <BE0131C6-2994-11D8-B146-000A95718F82@w3.org>
To: www-html@w3.org


Le 06 déc. 2003, à 07:03, Lachlan Hunt a écrit :
>  I have to disagree with this idea that a block element is 
> fundamentally presentational.

The problem is not completely as simple as that. It's exactly about 
editing and not only viewing. We always focus on viewing HTML and not 
editing it.

We are using elements: The notion of blocks is something which is more 
presentational than structural.

A paragraph is a syntactic and unique semantics structure.

Exactly
*** Source: Webster's Revised Unabridged Dictionary (1913) ***
Paragraph \Par"a*graph\, n. [F. paragraphe, LL. paragraphus, fr.
    Gr. para`grafos (sc. grammh`) a line or stroke drawn in the
    margin, fr. paragra`fein to write beside; para` beside +
    gra`fein to write. See {Para-}, and {Graphic}, and cf.
    {Paraph}.]

    2. A distinct part of a discourse or writing; any section or
       subdivision of a writing or chapter which relates to a
       particular point, whether consisting of one or many
       sentences. The division is sometimes noted by the mark ?,
       but usually, by beginning the first sentence of the
       paragraph on a new line and at more than the usual
       distance from the margin.

The presentation of the paragraph is something else. you see it as a 
block because of the history of the publishing industry and writings. 
For example, at the Roman times. There was no punctuation at all, the 
text was flowing without discontinuity.

The notion of paragraph is a notion of an idea.

My point for semantics element like address, accronym, cite, 
(blockquote, q), etc. is that they do not belong to block or inline. 
When you are editing, you can put in a block section inside a paragraph 
or you might want to put them outside of it. AND !! you might want to 
move them where the context change.

It will mean that for an authoring tool, it will have to manage the 
fact to know that
	1) it is "q" for example that I'm moving out of a block
	2) that I have to transform it as a blockquote.

Which might likely break many CMS and interoperability between 
applications.


>  A paragraph is a section of text, which can contain various elements 
> such as *emphasis* new lines
> abbreviations, lines of <code>, and various others that are grouped 
> together, hopefully, in a way that is meaningful.

no. a paragraph is the expression of an idea. See above.

>  As humans, we need this kind of seperation to be able to logically 
> understand documents which, of course, is presentational, however this 
> does not change the fact that these sections need to be logically 
> seperated within the markup.


agreed. But it has nothing to do with addresses or citation for example.

> This is why block elements cannot be purely presentational.  
> Rectangular boxes cannot be shown to a visually impaired person, 
> however sections can be through, as I said before, longer pauses 
> aurally, or larger gaps between brail symbols or whatever.  The point 
> is. the semantic structure of a block, or section, can be conveyed in 
> many ways.  Visually, as a rectangular box is just one.


The problem is the HTML 4.01 specification which is mixing concepts.

"p" and "blockquote" are not the same nature, they are both block.
but "q" and "blockquote" are the same nature and one is block and the 
other inline.

It's why I would be much in favor that all purely semantics tags like 
blockquote (or quote), address, etc... do not belong to the 
block/inline model, but just to the semantics model.

You can still choose to use the display property as block or inline.

>  The distinction between <ol> an <ul> is not just presentational.  For 
> example the following lists only really make semantic sense as an 
> unordered list, or ordered list respectively:
> <p>When I go shopping, I need to buy
>    <ul>
>        <li>Bread</li>
>        <li>Milk</li>
>        <li>Cheese</li>
>    </ul>
> </p>
> (This could, admittidly, also be ordered, however it doesn't really 
> matter whether bread is bought before milk or not)

Semantically I agree... except that for the XML Infoset it's still an 
ordered list :(. There's no notion of bag like in RDF schema. But I 
agree.

>  The point is, that the distiction between ordered and unordered lists 
> is not just a presentational one.  Presentation is simply the way in 
> which we, as humans, percieve the structure.

Agreed for machines too.




-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***
Received on Monday, 8 December 2003 10:43:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:59 GMT