W3C home > Mailing lists > Public > www-archive@w3.org > December 2002

RE: Editorial comments for Part 1, section 1-3

From: Henrik Frystyk Nielsen <henrikn@microsoft.com>
Date: Wed, 18 Dec 2002 18:47:00 -0800
Message-ID: <68B95AA1648D1840AB0083CC63E57AD60A1B3DA6@red-msg-06.redmond.corp.microsoft.com>
To: <noah_mendelsohn@us.ibm.com>
Cc: "Anish Karmarkar" <Anish.Karmarkar@oracle.com>, "Nilo Mitra" <EUSNILM@am1.ericsson.se>, "David Fallside" <fallside@us.ibm.com>, "Marc Hadley" <marc.hadley@sun.com>, "Martin Gudgin" <mgudgin@microsoft.com>, "Jean-Jacques Moreau" <moreau@crf.canon.fr>, <www-archive@w3.org>
Looking at the Namespaces 1.1 Candidate Rec [1], it does seem be a more
consistent use of the terms qualified name and expanded name. Here it
certainly looks like qualified name is {prefix, local}:

The  <http://www.w3.org/TR/xml-names11/#NT-Prefix> Prefix provides the
<http://www.w3.org/TR/xml-names11/#dt-prefix> namespace prefix part of
the qualified name, and must be associated with a namespace IRI
reference in a  <http://www.w3.org/TR/xml-names11/#dt-NSDecl> namespace
declaration. [Definition: The
<http://www.w3.org/TR/xml-names11/#NT-LocalPart> LocalPart provides the
local part of the qualified name.] 

 whereas expanded name is {iri, local}:

[Definition: An expanded name is a pair consisting of a
<http://www.w3.org/TR/xml-names11/#dt-NSName> namespace name and a
<http://www.w3.org/TR/xml-names11/#dt-localname> local name. ]
[Definition: For a name in a namespace, the namespace name is the IRI
identifying the namespace. For a name that is not in a namespace, the
namespace name is null. ] [Definition: The local name is the name
itself. ] It is this combination of the universally managed IRI
namespace with the vocabulary's local names that is effective in
avoiding name clashes. 

Interestingly, it refers to IRIs for which there is no formal definition
(yet) and I am not quite sure what the timeline is for that. While this
certainly seems to drift in the direction you indicate, this would be a
good thing to get coordinated during CR for both documents! As you know,
my usual concern is that it is dangerous to (re)define terms that are
part of other specs because it invariably leads to mutations.
 
Henrik
 
[1] http://www.w3.org/TR/xml-names11/#dt-expname

-----Original Message-----
From: noah_mendelsohn@us.ibm.com [mailto:noah_mendelsohn@us.ibm.com] 
Sent: Wednesday, December 18, 2002 17:47
To: Henrik Frystyk Nielsen
Cc: Anish Karmarkar; Nilo Mitra; David Fallside; Marc Hadley; Martin
Gudgin; Jean-Jacques Moreau; www-archive@w3.org
Subject: RE: Editorial comments for Part 1, section 1-3



Well, what the Namespaces Rec also says is: 

3. Qualified Names 

[Definition:] In XML documents conforming to this specification, some
names (constructs corresponding to the nonterminal
<http://www.w3.org/TR/REC-xml#NT-Name> Name) may be given as qualified
names, defined as follows: 




Qualified Name 




[6]  	QName 	::= 	(
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-Prefix> Prefix
':')?  <http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-LocalPart>
LocalPart 

[7]  	Prefix 	::=
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-NCName> NCName 

[8]  	LocalPart 	::=
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-NCName> NCName



The  <http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-Prefix>
Prefix provides the
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#dt-prefix> namespace
prefix part of the qualified name, and must be associated with a
namespace URI reference in a
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#dt-NSDecl> namespace
declaration. [Definition:] The
<http://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-LocalPart>
LocalPart provides the local part of the qualified name. 


Note that the prefix functions only as a placeholder for a namespace
name. Applications should use the namespace name, not the prefix, in
constructing names whose scope extends beyond the containing document. 



Although in Schemas we used the term Qualified Name to mean {uri,local},
I later discovered that the Namespaces rec. comes much closer to using
it for {prefix, local}.  See above.  The recommendation further
introduces the term "Expanded {Element,Attribute} Name" [1], but uses it
specifically in the form of an element that carries the uri and local
part.   



So, bottom line is:  while I personally like the term qualified name,
I'm not sure it's consistent with the terminology in the Namespaces
Recommendation.  Hence my suggestion that we at least introduce our use
of it.   


Noah 


[1] http://www.w3.org/TR/1999/REC-xml-names-19990114/#ns-expnames

------------------------------------------------------------------
Noah Mendelsohn                              Voice: 1-617-693-4036
IBM Corporation                                Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------







	"Henrik Frystyk Nielsen" <henrikn@microsoft.com> 


12/16/2002 12:39 AM 


        
        To:        <noah_mendelsohn@us.ibm.com> 
        cc:        "Anish Karmarkar" <Anish.Karmarkar@oracle.com>, "Nilo
Mitra" <EUSNILM@am1.ericsson.se>, "David Fallside"
<fallside@us.ibm.com>, "Marc Hadley" <marc.hadley@sun.com>, "Martin
Gudgin" <mgudgin@microsoft.com>, "Jean-Jacques Moreau"
<moreau@crf.canon.fr>, <www-archive@w3.org> 
        Subject:        RE: Editorial comments for Part 1, section 1-3



Hmm, I have always read the term "XML qualified name" in the context
stated in the description in the namespace spec:

"Note that the prefix functions only as a placeholder for a namespace
name. Applications should use the namespace name, not the prefix, in
constructing names whose scope extends beyond the containing document."

I vaquely remember that the WG agreed on using this term some time ago
but I can't find the reference anywhere. Can somebody remember the
details?

The potential problem by using the old phrase is that it is not clear
whether or how the "combination" should be computed: because we don't
say: should they be concatinated, or something else?

In any case, we use the term "XML qualified name" fairly consistently so
if we have problems in section 2.4, then we may have it in other places
as well. In any case, I agree that it is an editorial edit.

Henrik

________________________________

From:                  noah_mendelsohn@us.ibm.com
[mailto:noah_mendelsohn@us.ibm.com]                 
Sent:                  Fri 13-Dec-02 20:59                 
To:                  Henrik Frystyk Nielsen                 
Cc:                  Anish Karmarkar; Nilo Mitra; David Fallside; Marc
Hadley; Martin Gudgin; Jean-Jacques Moreau; www-archive@w3.org

Subject:                  Re: Editorial comments for Part 1, section 1-3

                 

I know I'm late going through these, but I think it's appropriate to
continue editorial refinements as we move between CR and PR.   Overall,
I
think your proposed improvements are terrific.  There's one I would roll
back (presuming it happened:  I'm offline and can't get to the latest CR
draft):

You propose:

* S2.4, P1: Change "by the combination of [local name] and [namespace]"
to "by the XML qualified name"

Though I like this use of "qualified name" in principle, I don't think
we've introduced it in a rigorous manner, have we?    Lacking that, I
think the original is less ambiguous.  Some might read the revised as
applying to the {prefix,lname} pair.  I'd revert to the original.
Thanks.

------------------------------------------------------------------
Noah Mendelsohn                              Voice: 1-617-693-4036
IBM Corporation                                Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------
Received on Wednesday, 18 December 2002 21:47:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 14:17:24 GMT