Comments on Section 2.6 of XML-Namespaces

[james anderson]

Andrew Layman wrote:

> The points you raise were addressed during the discussion leading up to the
> current paper.  Briefly, the namespaces facility does not address how
> "global" attributes are to be declared. It says merely that if a single
> attribute is used in two different element types, it may use namespace
> qualification to represent that it is in fact the same attribute, not two
> attributes with equivocal names.
>
> Nor does the namespace facility express an opinion on when such use is
> reasonable. As you note, the XML-Link proposal has found it useful.
>
>         I hope this summary is useful.

not yet. as it stands, this last statement, especially, confuses me more than it
helps. in the subjunktive it would be ok. the xml-link document itself proposes
to contend with ambiguous attribute names through name remapping. (at least in
the released version "www.w3.org/TR/WD-xlink#remapping".) for which one does not
need "namespaces". (or rather attribute remapping is itself a form of
ideosyncratic support for namespaces (please see the thread on namespaces and
their bad rep)).
with support for namespaces (even the form of conventional "namespace" invoked
with comments and "dotted" names, that is the form where the element/attribute
definer gets to say what is "theirs") you do not need to map attributes.
according to that way of working, if the application element uses the qualified
element name, then that's what it means. (nb. i do not claim this form of
namespace support is sufficient, i describe it to illustrate the point only.)
the example from WD-xlink would simply look like:

<TEXT-BOOK title="Compilers: Principles, Techniques, and Tools"
             role="PRIMARY" xml:title="Primary Textbook for the Course"
             xml:role="ONLINE-PURCHASE"
             href="/cgi/auth-search?q="+Aho+Sethi+Ullman"/>

you don't need a pi in the application document for this, you just need some
comments in the source schema and a registry in which authors stake claims to
the names.
while xml-link would benefit from namespaces, it doesn't yet appear to show it.


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)