W3C home > Mailing lists > Public > www-html-editor@w3.org > July to September 2004

[XHTML2] Some comments

From: Anne van Kesteren <fora@annevankesteren.nl>
Date: Thu, 29 Jul 2004 14:12:16 +0200
Message-ID: <4108E9A0.80704@annevankesteren.nl>
To: www-html-editor@w3.org

Comments on the Working Draft of 22 July 2004[1]


1.1.1. Design Aims

# As generic XML as possible: if a facility exists in XML, try to use
# that rather than duplicating it.

If that is true, why are several elements imported from XHTML 1.1? For 
example: CODE, VAR, BLOCKQUOTE et cetera.


1.1.2. Backwards compatibility

# Much of XHTML 2 works already in existing browsers;

This is not true. Even in Mozilla I would have to include the default 
style sheet in order to make it work as described. And if XHTML 2.0 will 
get a new mime type, like 'application/xhtml-2+xml' user agents would 
ask to download it instead.


1.2. Major Differences with XHTML 1

I think that BR can still serve a useful purpose, when you don't want to 
split text in lines, but just use a line-break. Example:

  <p>John,<br/>
  ...</p>

"John" is not a separate "line of text". I actually think L is 
presentational, although in some rare cases I never encountered it can 
probably add some semantics.


8.1. The address element

There are a lot of use cases where it would be nice if this could be an 
inline element as well. As in having both BLOCKADDRESS and ADDRESS.
Having more elements allowed within (BLOCK)ADDRESS, such as headings and 
paragraph is a good thing for more detailed contact information imo.


8.4. The div element

It looks like DIV elements were used for presentation in the example. 
Aren't they just marking up sections of the document? (So remove that 
example, since there is SECTION element already.)

Also, in the same example a P element is used:

# <p>By Huntington B. Snark</p>

Should I call tag abuse? ;-)


8.5. The heading elements

It is confusing that there are two types of headings. Is there an 
example were there is need for both? Also, is the following considered 
bad practice:

   <section>
    <h>...</h>
    <section>
     <h1>...</h1>
    </section>
   </section>

Could the draft clarify that? And how they should interact with each other?


8.6. The p element

In the example here, both PCDATA and block level elements are mixed. 
This looks really bad and just asks for markup abuse once released to 
the public.


8.7. The pre element

The example is a quote, why is PRE used? And not BLOCKQUOTE with the 
style rule mentioned in the description?


8.8. The section element

The example given here is almost exactly the same as in 8.5.


8.9. The separator element

There isn't mentioned a single use case. Only some presentational issues 
that should be kept in CSS as mentioned in 1.1.1. Right?


9.1. The abbr element

Should this element get it's own attribute, instead of abusing the TITLE 
attribute for it's purpose?


9.2. The cite element

The examples make it look "Gandalf the White" and "[XML]" are citations. 
(There is also an error in the first example.) Shouldn't the HREF 
instead of the CITE attribute be used?


9.6. The kbd element

How should keys be marked up, or is this not the right attribute for 
that? Use case:

  To go to the homepage, click ALT+1

How should "ALT+1" be marked up? I don't want someone to type in "A+L+T+1".


9.7. The l element

Didn't BLOCKCODE preserve whitespace by default? What do we need the L 
element for here? And as mentioned before, BR is still needed. I also 
think that a better use case for L should be presented, this one is bad.


9.8. The quote element

# either directly in the text, or via a stylesheet

Why directly in the text, what is the use case?


10.1. The a element

Why do we need this element when there is SPAN? Everything can be an 
explicit link in XHTML 2.0.


11. XHTML List Module

Why can't LABEL be used halfway in a list? Example:

  <nl>
   <label>...</label>
   <li>...</li>
   <li>...</li>
   <label>...</label>
   <li>...</li>
  </nl>

I also don't really like how nested lists work. I have read something[2] 
in the past that seems a whole lot better. Maybe someone can look into it?


12.1. Core Attribute Collection

Will ID be replaced with it's XML equivalent, once that specification[3] 
becomes a recommendation?

I think something should be done with TITLE. On www-html there is a 
permathread[4] going on with some suggestions. Suggestions include to 
replace it with an element, create more attributes to be able to 
distinguish between titles, practical hints and descriptions.


13.1. Hypertext Attribute Collection

Why are HREFLANG and HREFTYPE needed at all? It currently looks like 
they define what kind of content you are going to see when you follow a 
link. Let's see that resource http://example.org/x is written in 5 
languages and 3 different types. A friend of my creates a link to it:

  <a href="http://example.org/x"
     hreflang="nl"
     hreftype="text/html"
  >...</a>

Now I follow that link and obviously, since I'm browsing a conforming 
browser I get the Dutch HTML version of that page. Now I'm passing that 
link along (copying it from my address bar) through IRC and tell him 
it's a great Dutch site (what do I know?).

A minute later he tells me he can't read Russian and why I think giving 
him that link is funny.

If a document consists of multiple versions it should have a permanent 
link (cool URIs don't change) for each document. That would mean 3*5=15 
different documents for the above given example each with their own URI 
possibly looking like: http://example.org/x.nl.html

Also, isn't TARGET a CSS issue[5]?


15.1. Bi-directional Text Collection

The second example here does define a DOCTYPE, but doesn't define a 
namespace.


16.1. Edit Collection

Why does |edit="deleted"| has default presentation where QUOTE does not?


17. XHTML Embedding Attributes Module

I can't see why modern user agents need TYPE. Most of them don't need it 
with HTML anyway.


18. XHTML Image Map Attributes Module

Why aren't the attributes directly applied to the NL element? Using a P 
element there looks like a major hack.

And how do I describe sections that are not links and are not part of 
the navigation, but are part of the same image? I guess not with LI, 
since that would be a hack, although it is suggested in the specification.


21.1. The object element

Why do we need this?


23. XHTML Scripting Module

Since this is getting dropped, is the W3C working on something along the 
lines of the following:

  <?xml-script href="script"
               type="application/x-javascript"
               alternate="yes"
  ?>

...?


24. XHTML Style Attribute Module

Can this be dropped as well? Having a STYLE element is bad enough.


27. XForms Module

Can this be replaced by understandable for everyone forms?


This was it (besides the previous mails sent). I probably missed things.


[1]<http://www.w3.org/TR/2004/WD-xhtml2-20040722/>
[2]<http://webperso.easyconnect.fr/danielglazman/weblog/newarchive/2002_11_24_glazblogarc.html#s85072264>
[3]<http://www.w3.org/TR/2004/WD-xml-id-20040407/>
[4]<http://lists.w3.org/Archives/Public/www-html/2004Jul/thread.html#16>
[5]<http://www.w3.org/TR/2004/WD-css3-hyperlinks-20040224/>


-- 
  Anne van Kesteren
  <http://annevankesteren.nl/>
Received on Thursday, 29 July 2004 08:13:10 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:52 UTC