Re: Namespace Identifiers - URI, URN, URL?

[Michael Beddow]> Sorry about the previous empty posting!

> So the original posting ought to have read:
> "Just be sure your document uses  UTF-8 encoding if you
> don't put it, because that's the default encoding..."
> then continued:
> "Any bytes in your XML  document that are not part of a valid
> utf-8 encoding sequence will cause your XML document to become invalid"
> and concluded with:
> "It's better to put the XML declaration in and explicitly state the
> encoding used"
>

This makes it seem as if there might be a difference between declaring UTF-8
encoding and omitting the encoding declaration (although actually there
wouldn't be a difference).  The problem you mentioned about declaring
iso-8859-1 but actually including non-ascii characters is just as likely to
happen in reverse with many computers, especially with older operating
systems or applications.

The real trouble is that a person can't easily determine what the true - as
opposed to the claimed or imagined - encoding really is.  Until this is
possible, the encoding declaration can still turn out the be inaccurate.  It
seems we need a document checker that can determine if a document's actual
encoding is consistent with some particular declaration.

As a nit, aren't encoding problems a lexical  issue (as opposed to a
validity issue)?  They get discovered and handled (or not) even before
well-formedness checking, don't they?

Cheers,

Tom P


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list