Re: still too early for XHTML :-((

See comments below...

At 11:08 PM 4/7/2003 +0200, you wrote:
>On Po, 2003-04-07 at 19:37, Jacob Kjome wrote:
> > So, can anyone confirm that if an XHTMLScriptElement is added to the
> > XHTML DOM that it is written as <script></script> instead of <script
>
>I can confirm that it's written out as <script /> and while it is a
>problem for browsers with closed source code it is not a real 'bug'
>since W3C allows that.
>
> > If it is writing it as <script /> then we will have a major browser
> > incompatibility.
>
>yes, both IE 5-6 and Opera 7 choke on that and stop rendering the page.
>
> > It should definitely be written as <script></script>.
>
>But the <script /> is XHTML compatible. XMLC does that to all tags when
>generating XHTML. Perhaps the David Corbin's patch with configurable
>behaviour will help but then the question is why to bother with
>producing XHTML 1.0 when the output is basically equal to HTML 4.01.

When I say that it should be <script><script> and you say "but <script /> 
is XHTML compatible" we are arguing two different things.  I agree, <script 
/> is legal XML syntax and there is nothing in the XHTML1.0 spec that I can 
see that explicitly says it can't be used this way.  However, the HTML4.01 
spec, of which XHTML1.0 is an XML'ized version, declares that <script> tags 
are required to have an </script> end tag.  Other than having to follow 
various XML rules, the only real change between HTML4 and XHTML1.0 are that 
the Empty elements in HTML4 need to have " />" added to them and there is 
no more attribute minimization.  So, yes, XHTML1.0 *is* HTML4.01 with XML 
rules applied.  The benefit of this is all the benefits you get from 
XML.  Of course the XHTML1.1 spec goes further and allows for a lot more 
flexibility such as inventing your own elements with your own 
behaviors.  XHTML1.0 was a stepping stone meant to get people using XML 
syntax while still proving for backward compatibility with HTML.  And that 
is the key "backward compatibility".  If <script></script> and <script /> 
are both legal XHTML, but the latter is not backward compatible with 
browsers that only understand HTML, then we should be using 
<script></script>.   We have everything to gain and nothing to lose.  Where 
is the harm?  This one's a no-brainer.

>Let me list all the XHTML problems I encountered so far, IIRC:
>
>1) document type incorrect, javascript didn't work
>2) too old DTD for XHTML in XMLC

Is this an XMLC issue or an Enhydra framework issue?  How do you set the 
doctype?  i set mine like this...

first, grab a reference to the default output options of the document and 
then do...

oo.setOmitDocType(false);
oo.setPublicId("-//W3C//DTD HTML 4.01 Transitional//EN");
oo.setSystemId("http://www.w3.org/TR/html401/loose.dtd");

Then I pass the output options object to the dom writer.

XMLC doesn't do this automatically, so how could you get the wrong doctype 
when you set it yourself?...which is why I think this is an Enhydra issue, 
not an XMLC issue.  Maybe you can help me understand?

>3) setting of boolean attributes does not work

Yep, definitely need this patched!  Did you provide this patch (and others) 
to David or Richard?  I could also apply them, but I'd feel much more 
comfortable having one of them do it.  Maybe provide a .zip file with all 
gathered patches included? and send to the list on this thread.

>4) single quotes are converted to &apos; (IE doesn't understand)

Like I said in my email last night, the fix for this is to use &#39; 
instead of &apos; since the latter only appeared in XML 1.0 and not 
HTML.  I tested the following and it worked perfectly in both Mozilla and 
IE6...

<a href="#" onmouseover="alert(&#39;hello, world!&#39;);">Hello World</a>

Again, since both entities are perfectly compatible with XHTML, but only 
&#39; works with browsers that only understand HTML, we should use the 
backward compatible one.  Again, we have everything to gain and nothing to 
lose by doing this.

See: http://www.w3.org/TR/xhtml1/#C_16

>5) <script /> is not understood by IE/Opera

Already addressed this above.  We should definitely use <script></script>.

>6) URLRewriter does not work

What was the issue with this one?  I don't remember.

>I reported all problems and submitted patch for 1 and 2 and suggested
>solutions for 3 and 6. David C. submitted patch for 4 and 5.
>
>Even if the 3 got fixed (not that hard) the 6 is still a serious issue.
>
>Also note that XMLC with patch 5 basically produces HTML 4.01 code,
>since the only visible difference between HTML and XHTML is the short
>form of <TAG />. So after disabling the short form of closing tag the
>question is why to bother with XHTML at all.

http://www.w3.org/TR/xhtml1/#why

And that is just for XHTML1.0.  XHTML1.1 is even more flexible, although 
less backward compatble and should only be used when you know your audience 
are using browsers which are XHTML compatible.  Most are not.  Mozilla 
mostly is, but even it has a few issues with strict XHTML (when served with 
the application/xml+xhtml or text/xml content types).

>Petr
>


Jake

>_______________________________________________
>XMLC mailing list
>[email protected]
>http://www.enhydra.org/mailman/listinfo.cgi/xmlc