Re: Names beginning with "xml"

On Wed, Aug 15, 2012 at 10:37 PM, James Clark <jjc@jclark.com> wrote:

> For elements and attributes, I would say
>
> - documents MUST NOT use an attribute name "xmlns" , assuming we go
> for a no prefixes in names options; this would not be strictly
> required by the no-prefix decision, but if we go for no prefixes, and
> things it's best to exclude XML Namespaces entirely
>
> - documents SHOULD NOT use element/attribute names starting with xml
> (modulo whatever we decide on the "xml:" prefix)
>
> - processors MUST accept element/attribute names starting with xml,
> other than an "xmlns" attribute
>

-1, particularly on this third part. I've already admitted my biggest bias
is that I use the restriction against names starting with "xml" as an
important part of Amara's generic, schema-free data binding mechanism, but
of course that's not enough reason in itself.

This would also provide a very large opening for creating MicroXML
documents which are not backwards compatible with XML 1.0.

My guess is that your idea here is to get MicroXML out of the business of
deciding what "xml..." names are kosher. Just say "we'll allow it except
for xmlns."  My counter-proposal would be quite similar to yours in
implementation.

* The processor must report all elements and attributes beginning with
"xml", however, "xmlns" attributes are not permitted. Attributes beginning
with "xml" are reserved for XML core specifications.

That last sentence (edited into proper spec speak) is the difference.  I
would want MicroXML to say, as XML 1.0 does, that people should not create
their own "xml..." elements and attributes.


-- 
Uche Ogbuji                       http://uche.ogbuji.net
Founding Partner, Zepheira        http://zepheira.com
http://wearekin.org
http://www.thenervousbreakdown.com/author/uogbuji/
http://copia.ogbuji.net
http://www.linkedin.com/in/ucheogbuji
http://twitter.com/uogbuji