|
|
Posted by Alan J. Flavell on 01/12/10 11:49
On Mon, 5 Jun 2006, Rik wrote:
> Alan J. Flavell wrote:
> > On Mon, 5 Jun 2006, Rik wrote:
> >
> >> <meta http-equiv="Content-Type" content="application/xhtml+xml;
> >> charset=ISO-8859-1" />
> >
> > Complete nonsense, I'm afraid.
[...]
> I was trying to post a few usefull comments in a minimal amount of
> time, but OK, here it is.
> http://www.w3.org/TR/xhtml-media-types/xhtml-media-types.xhtml#media-types
I'm afraid you're missing the point.
> While the content type IS application/xhtml+xml, you COULD use
> text/html, for a good reason:
If you have provided a content-type of text/html from the server, then
a <meta http-equiv...> is far too late to be trying to change your
mind and say that it's really application/xhtml+xml
On the other hand, if you have provided a content-type of
application/xhtml+xml, then the rules for specifying the character
encoding will already have done their work and fixed the character
encoding, long before it gets to the point of seeing that bogus <meta
http-equiv...> that's possibly trying to make it be something else.
> text/html:
> Appendix C "HTML Compatibility Guidelines"
....for XHTML/1.0 only...
> summarizes "design guidelines for authors who wish their XHTML
> documents to render on existing HTML user agents".
We've done this part so often before, I'm not going to repeat the
arguments all over again; but they really aren't part of what it takes
to understand that:
> >> <meta http-equiv="Content-Type" content="application/xhtml+xml;
> >> charset=ISO-8859-1" />
is nonsense.
> That was the nagging in the back of my head. And now I will further
> prove to you I'm lazy: I will NOT investigate which browsers are
> XHTML compliant, and which aren't.
That wasn't any part of the specific problem, either.
*Even* if you could find a client implementation which accepted
text/html from the server, and changed its mind about content-type and
character encoding when it saw this ominous <meta http-equiv...> , it
would still be complete nonsense in terms of the WWW specifications.
[Back to original message]
|