XML, NAMESPACE and HREF ?

[Peter Murray-Rust]

At 13:18 23/01/98 -0400, David G. Durand wrote:
>At 11:50 AM -0400 1/23/98, Patrice Bonhomme wrote:
>>Hi,
>>
>>I am working with XML and are interresting in using some NAMESPACE. But i
was
>>wondering about the content and the organization of the "NAME SPACE" on the
>>server side ? What is there behind the HREF in :
>>
>><?XML:NAMESPACE HREF="http://www.blahblah.fr/mydtd/" AS="BLAH"?>

It's highly likely that the keywords will end up in lower case so that it's
probably worth writing:

<?xml:namespace href="http://www.blahblah.fr/mydtd/" as="BLAH"?>

The xml:* case-sensitivity is already determined for its use in XML and (I
think it was EveM ) a recent authority suggested that XLL would probably
use lower case.

The BLAH depends on *your* case sensitivity of course :-)

>>
>>Could it be an XML-DATA specification (Schema ?) for the DTD ?

This is how I use it in Jumbo. The href points to an XML file with the
equivalent of a DTD, and a lot of additional material for each namespace and 
many of the elements (on a per-element basis).  Using XML simplifies the
processing over a DTD.  But, as DavidD correctly says, there is no
standard. The advantage of using XML syntax in this file is that it will
probably be very easy to transform it to whatever emerges in the future (if
anything).

IMO XML-DATA has several themes:
	- the use of XML syntax for DTD-like material
	- the addition of more material (e.g. data typing) to that material
	- the creation of additional data structure (relations)

I find the first two straightforward and valuable and have essentially
implemented them in Jumbo. The third is more difficult and has already
given rise to much discussion on this list :-)


>
>Just decide what you want and implement it. Namespaces are not standardized
>yet, and are still the focus of some debate. Some. like me, believe that
>they are syntactic sugar for things you can already do in XML with
>attributes. The issue of what the URL points to is currently entirely open.
>The only guarantee is that the absolute URL, plus the part of the element
>name _after the colon_ is uniquely associated (by unspecified means) with
>some definition of semantics. The putative purpose of namespaces is to
>provide applications that might need them with globally unique IDs for
>element types.
>
>They are an additional layer on top of XML and don't affect XML syntax.
>They are also experimental, and not standard at the moment.

Just to confirm that I agree with the fundamentals of what David has said.
Personally I find namespaces useful because they are more intuitive to me
that other approaches such as architectural forms, but I am sure that many
people find just the opposite.  I think that they are most useful when
there is a well-defined global role for the namespace (e.g. MathML:, HTML:)
as it improves the human readability of the document and helps define where
to locate per-element processing (if stylesheets are inappropriate). 

	P.

Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)