Re: New requirement - Simplifying Meta Data Profile

Lachlan Hunt schreef:
>>> So, yes, XHTML 2 threw out backward compatibility. 
>>
>> I think this is a common mistake, and given the charter of XHTML2 
>> maybe not a strange one. However, this does not reflect the reality 
>> where XHTML2 contained elements like <h1>...<h6>, <a>, <img>, and 
>> also re-uses the XHTML1 namespace.
>
> I'm yet to see a draft of XHTML 2.0 that reuses the XHTML 1.0 
> namespace.  AFAIK, that intention was only mentioned once in an email 
> on www-html-editor.  Perhaps it is/was in some internal draft, but 
> hopefully it will never make it into any published spec.

I looked it up, you’re right, I don’t think it was actually resolved. 
For reference: 
http://lists.w3.org/Archives/Public/www-html-editor/2005AprJun/0154.html

I received this email in reference to that message:

   HTML Working Group XHTML 2 Issue Tracking System notification

   shane moved PR#7808 from Conformance to 6
   Message summary for PR#7808
       From: Laurens Holst <lholst@students.cs.uu.nl>
       Subject: Change XHTML 2.0 namespace to http://www.w3.org/1999/xhtml
       Date: Thu, 02 Jun 2005 19:18:16  0900 (JST)
       0 replies     0 followups
       Notes: We plan to do this.

Note the last line. That’s all.

>> The existence of these elements and this namespace can only be 
>> explained as being for backwards compatibility reasons,
>
> While it may appear from the tag names that the intention is for 
> backwards compatibility, that is not true.  These are just some of the 
> incompatibilities introduced in XHTML 2.0, some of which would make it 
> totally impossible to implement in the real world:
>
> 1. <script> renamed to <handler>
> 2. <input> has been replaced with the XForms version in the XHTML1
>   namespace, which is incompatible with the XHTML1 version.

Sorry, I should have clarified. When I said XHTML2 documents I meant 
real documents (as-in text; a manual, a review, a research paper, etc). 
No forms, no script, just plain marked up text. Those are ‘backwards 
compatible’ (somewhat) in XHTML2, under certain conditions. For forms it 
would be pretty much impossible, as XForms does not resemble HTML forms 
at all. Same is valid for <label> below.

> 3. <img> drops alt in favour of nested fallback content

Works in XHTML1 browser, because it’s XML. Of course, only visually.

> 4. <object src=""> instead of <object data="">

Granted. But there’s <img>, so that catches most, fortunately.

> 5. href, src, etc. on every element is, according to browser vendors,
>    extremely difficult or impossible to implement in the real world.

This item is moot in this list; that was exactly why I mentioned <a> and 
<img> were retained. Also, whether it is implementable doesn’t have very 
much to do with backwards compatibility I think.

Anyway, I don’t see why it would be impossible. Attach an on click 
handler to every element with href which changes the location on click. 
The presence of an src attribute will not show the element and its 
contents, and instead replace it with the referenced resource. 
Relatively easy to do with XBL.

> 6. <label> used for labelling lists, not form controls
> 7. <hr/> renamed to <separator/>
> 8. <l></l> replaces <br/>

<separator> and <l> can very easily be gotten to work with a few 
stylesheet rules. l { display: block; } and separator { display: block; 
height: 1px; border: 1px outset gray; padding: 1em 0; }… something like 
that. Add some XBL to the mix, and you can fix points 1, 3, 4 and 5 too.

Finally, I’m not saying it’s perfect (note: the spec isn’t finished), or 
that it takes the same approach as HTML5. Because it is not in the 
charter I suppose it is also hard to allocate resources to it, to 
justify the retaining of certain elements, and to do it really 
consistently and thoroughly, so maybe it should not have been done in 
the first place. However, because there are other problems with XHTML2 
does not mean that ‘it is not true’. I think it *does* show the 
intention. Otherwise, why have these things. There would be no point. 
Explain to me the other plausible explanation.

Anyway, just wanted to add some nuance to that one sentence “So, yes, 
XHTML 2 threw out backward compatibility.” On one hand, yes, on the 
other hand, not as entirely as it might appear.


~Grauw

-- 
Ushiko-san! Kimi wa doushite, Ushiko-san nan da!!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Laurens Holst, student, university of Utrecht, the Netherlands.
Website: www.grauw.nl. Backbase employee; www.backbase.com.