Re: Validator doesn't send HTTP_ACCEPT headers, "Conflict between Mime Type and Document Type" warning is incorrect.

Am 31.07.2007 um 16:20 schrieb Andreas Prilop:

> Hint:   <BR>   <br />
> If you really don't understand the difference between HTML and XHTML,
> I won't explain it to you.

I know the difference very well, Andreas. :-)
Maybe you should read the Specs concerning that point (http:// 
www.w3.org/TR/xhtml1/#guidelines), again (and the way they are  
handled by the browser's parser).

>> As you proved above with your own documents, you serve it twice. And,
>> more difficult, you have to provide it twice on the server.
>
> http://www.unics.uni-hannover.de/nhtcapri/ruby-annotation.html
> http://www.unics.uni-hannover.de/nhtcapri/ruby-annotation.xhtml
> exist twice because ruby markup formally exists in XHTML 1.1 only.
> But I have an (unspecified) HTML version, too, to show that
> Internet Explorer 6.0 will also display ruby markup with complex
> scripts.

Andreas, please raise your head from your own single test files, and  
have a look into the real world out there!
If you are able to provide me a solution to *dynamically* distinguish  
(and set the apropriate Mimetype) and to *avoid* storing the files  
*twice*, than you would be my hero!

>
>> But what about dynamically served and maybe frequently
>> changed documents? What about documents generated by a Content
>> Management System or Blog?
>
> If you want to serve content on the web, do it in HTML 4 (Strict)
> and forget about XHTML! The web runs on HTML.

And this suggestion on a W3C mailinglist... :-)
Would you discuss this within the working groups, please? :-)

>> You have to provide each document twice as .html document and
>> as .xhtml document?
>
> Of course not! When you want to present come document, do it with
> HTML 4 and text/html !
>
> It's only you, who has the silly obsession to serve one and the same
> document with different MIME types. What's the point?

The point is, that the XHTML Specs do allow that and stronger,  
*force* the use of "application/xhtml+xml" over "text/html" ...
What I only try, Andreas, is stick to the XHTML Specs and take as  
much out of them as possible. And therefore, you unfortunately *have*  
to distinguish between the browser's capabilities.
If you don't do that, you totally have either to lock out Internet  
Explorer, or you have to avoid any XHMTL, because even XHTML 1.0  
Strict says, to prefer "application/xhtml+xml" over "text/html".  
Thers is a SHOULD over a MAY, Andreas.
If we follow your argumantation and your advices, there *must* not be  
*any* XHTML-1.0 based website out there, because, if you stick to the  
words of the Specs, you should prefer to serve XHTML documents as  
"application/xhtml+xml" rather than as "text/html".
If you serve the latter, you indeed don't need to serve XHTML anyway,  
in that case you can stick to good old HTML 4. And if you *want* to  
use XHTML? If your content suffices all to serve it as XHTML?
So the purpose, and that you can also read within the lines of the  
specs, is to give the client the best of what it is able to render.  
This purpose isn't new, it is the red line of HTML at all: serve the  
best, any asking client accepts. For XHTML-capable browsers this  
would be a XHTML Mimetype, and for NON-XHTML-capable browsers it  
could be text/html.

> *You* have an obsession to serve one MIME type to Internet Explorer
> and another MIME type to other browsers. I continue to call this
> a stupid idea - no matter what olivier Thereaux says about politeness.

I call this to take the idea and the approaches of the W3C and its  
Specs for granted, Andreas.
I try to take the leading banner of the W3C (http://www.w3.org/) for  
serious: "Leading the Web to Its Full Potential..."

-- 
Sierk Bornemann
email:            sierkb@gmx.de
WWW:              http://sierkbornemann.de/

Received on Tuesday, 31 July 2007 15:02:11 UTC