W3C complience checker and conformance

[Toby Inkster]

So THAT'S what that's about. I've just noticed recently that the W3
validator is insisting on a character encoding declaration. Something like:
<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1" />
in order for it to be strict XHTML.

A character encoding most be declared for the validator, but you don't
need to use a META element to do so.

 

[ironcorona]

http://www.w3.org/TR/xhtml-media-types/

Beautiful. Thanks.

 

[ironcorona]

A character encoding most be declared for the validator, but you don't
need to use a META element to do so.

Perhaps. It's quite new to me so when I didn't have one I followed the
link it showed:
http://validator.w3.org/docs/help.html#faq-charset

I followed this about and it suggested the meta tag.

 

[Spartanicus]

So would you care to show some sort of reference as to why it is
"incorrect XHTML" to use a closing tag on an empty element, since the
part of the specs I quoted says that it *is* correct?

I shouldn't have suggested that it was a spec compliance issue. Although
there isn't even a "SHOULD NOT" recommendation in the spec, and the
informative note you quoted even suggests the opposite, I consider using
a closing tag for empty elements poor practice, on par with quick
closing non empty elements, which IIRC is expressed as a "SHOULD NOT" in
the spec.

 

[Alan J. Flavell]

Perhaps. It's quite new to me so when I didn't have one I followed
the link it showed:
http://validator.w3.org/docs/help.html#faq-charset

Indeed:

| Specifying a character encoding is normally done in the web server
| configuration file [...]

I followed this about and it suggested the meta tag.

I'm not sure how you managed to read that into it!

meta http-equiv is *only* relevant to HTML (and consequently to
Appendix-C XHTML/1.0 being offered as text/html, since that is meant
to be handled as error-fuxup-ed HTML).

It has no meaning for XHTML-properly-so-called.

It's a workaround which can be used (for HTML, not XHTML) for
protocols (such as file: or ftp which do not include a mechanism for
carrying a character encoding specification (MIME charset) in the
protocol. But for HTTP, the better place to deal with this is in the
HTTP header, just as the W3C FAQ says - by suitably configuring the
server.

The cited WDG tutorial (which is primarily addressing HTML, not XHTML)
says the same:

|An HTML document must specify its character encoding. The preferred
|method of indicating the encoding is by using the charset parameter
|of the Content-Type HTTP header.

 

[Mark Parnell]

Deciding to do something for the good of humanity, Spartanicus

I shouldn't have suggested that it was a spec compliance issue. Although
there isn't even a "SHOULD NOT" recommendation in the spec, and the
informative note you quoted even suggests the opposite, I consider using
a closing tag for empty elements poor practice, on par with quick
closing non empty elements, which IIRC is expressed as a "SHOULD NOT" in
the spec.

Ah, we agree then. It is certainly bad practice.