Namespaces !

[James Clark]

Patrice Bonhomme wrote:

>  1/ The example following this definition uses a NS declared after its use :
>         "<html:html xmlns:html='http://www.w3.org/TR/REC-html40'>"

This is allowed.  The spec should be clearer about this.

>     I am not sure that attribute-based is the best way for declaring NS. Why
> not have preserved the old specification for the declarations of XML
> Namespaces (using PI) ?

This has been discussed an exhaustive length in the WG.  If you are a
W3C member, you can go read the archives.  I don't think it is
profitable to rehash the discussion in this forum.

>     Create another reserved name (xmlns)

The XML Recommendation already reserves xmlns and indeed all names
beginning with "xml".

> weighs down the XML notation and
> opens the door to already encountered problems (remenber HTML!). Each one
> (Microsoft, Netscape, Sun, ...) will arrive with its own reserved name and one
> will fall down in the same problems as with HTML (<OBJECT> vs <EMBED> for
> example). 

Only names beginning with "xml" (any case) are reserved and they are
reserved for use only by future versions of XML.

> We should have only one reserved name : "xml" !!!
> 
>  2/ There is a redundancy of information.
> 
> A simple prefix is enough to specify the namespace used :
> 
>         <book xmlns='urn:loc.gov:books' xmlns:isbn='urn:ISBN:0-395-36341-6'>
>           <title>Cheaper by the Dozen</title>
>           <isbn:number>1568491379</isbn:number> ...
> 
> Should be :
> 
>         <book xml:ns="books"> <!-- the default NS is "books" -->
>           <title>Cheaper by the Dozen</title>
>           <!-- switch to "isbn" NS -->
>           <isbn:number>1568491379</isbn:number> ...
> 
>     This makes also XML document not easily readable.

Where is the "isbn:" prefix declared?  How would that allow global
attributes with different namespaces?

You are wasting your time suggesting alternative designs.  Read the
"Status of this document" section: "the Working Group intends to keep
the features [the draft] describes functionally unchanged unless
problems are discovered during early implementation work".

>  3/ Implementation.
> 
> If i understand the new WD, it's possible to have everywhere within the
> document (in each Element start tag) a Namespace declaration. Hum, i agree
> with James Clark that it is easy to implement but we have to provide for each
> Element object an 'xmlns' attribute and make inherited each one of its
> descendants.

You have to be able to find the inherited value of namespace declaring
attributes; this is no different from what you already have to do for
xml:lang and xml:space.

James



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)