W3C home > Mailing lists > Public > www-dom@w3.org > January to March 2003

WD-DOM-Level-3-Core-20030226 : Appendix B: Namespaces Algorithm

From: james anderson <james.anderson@setf.de>
Date: Fri, 7 Mar 2003 00:45:51 +0100
To: www-dom@w3.org
Message-Id: <C371F10B-502D-11D7-AAE7-000393BB8814@setf.de>
B.1.1: Scope of a binding

"An element is said to be within the scope of the binding"

this phrasing is poor. to which binding does "the" refer? the reader 
may be expected to understand it to be "the binding of the element' 
identifier's prefix to the element's identifier's namespace name", but 
that concept is ungrounded at this point in the text.
the least one needs to do is to turn the sentence around, with which 
the "the" is more appropriately an "a":

"If an element's identifier's prefix is bound to the same namespace URI 
in the [in-scope namespaces] defined in [ XML Information set ] as the 
namespace URI of the identifier itself, then the element is said to be 
within the scope of a binding."

which calls attention to the root of the problem: that the concept 
"within the scope of the binding" is at odds with the definition of 
"the scope of a binding" as fixed in "namespaces in xml", according to 
which the inference flows from the other direction.

the consequence of the draft's inversion is that the it suggests that, 
where the apparent binding (after all there _is_ a binding) conflicts 
with the contingent binding which would be imputed from the incidental 
element identifier, it is necessary to introduce a new binding. this is 
not true. one need only change the prefix to reflect the actual 
apparent bindings. as a change to the attribute appears to be the 
approach suggested when "polishing" attributes, it is why does the same 
not apply to the element identifier?

...
Received on Friday, 7 March 2003 02:50:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:56 GMT