On Mon, Dec 13, 2010 at 9:35 AM, Amelia A Lewis <amyzing@t...> wrote:

On Mon, 13 Dec 2010 16:01:11 +0700, James Clark wrote:
> I wrote a post describing a fairly minimal subset of XML:
>
>    http://blog.jclark.com/2010/12/microxml.html

Nice.  I like the way that it copes with namespaces; I think it may
improve upon Michael Kay's suggestion, even.

I have a problem with it.

How do I tell whether it's safe to use my uXML parser instead of my
(heavier) XML 1.0 + Namespace in XML + XML:Base + XML:ID + whatever
parser?

XML declaration not allowed.  XML 1.0 and uXML are placed in the
position of both claiming to be the canonical format indicated by the
absence of a declaration.

I've only had a chance to skim ìXML (that would be a wikkid name :) ), but I intend to read more thoroughly when I get a longer break today.  I hope my comments are not complete rubbish as a result.

With what Liam and Henri said about magic numbers, and I agree, how about just:

<!DOCTYPE microxml>

No need for BOM since MicroXML is UTF-8 only, though I suspect there may be much discussion on that point.  The MicroXML decl *must* be at the beginning (I agree the need for magic number trumps convenience for cut&paste.  One can always add a step in the toolchain to fix misplaced microxml declarations.)

I think the above does not violate HTML5? (Not sure).  I understand that HTML compat rules out:

<?microxml?>