Why not PIs for namespace declarations?

[Clark C. Evans]

On Thu, 23 Dec 1999, Frank Boumphrey wrote:
> In fact if you look at the working Drafts you will see that 
> namespaces used PI'sright up to the final recomendation!

Yes, sounds like a fast change of heart.  The PI version even
made it into the XML Companion by Neil Bradley...

 <?xml:namespace name="http://www.vehicles.org/"
                 href="http://www.vehicles.org/dtds/v.dtd"
                 as="veh"
 ?>

> 1. My understanding is that they were abandoned to allow scoping.

Yes.  I had assumed (due to attributes being used) that 
PIs were only valid in the prolog... now that David pointed
out that this was a bad assumption -- I'm quite puzzled.

> 2. A mecanism to not scope attributes was wanted.

Is this related to that "three name space partitions" stuff?

> 3. there was a general feeling that PI's were 'broken'

Thanks Frank!

...

Now I'm wondering if it wouldn't be all that hard for parser's
to support both methods.  As the PI version seems far cleaner.

  <pre:element xmlns:pre="http//www.vehicles.org/dtds/v.dtd" />

becomes

  <?xml:namespace href="http://www.vechicles.org/dtds/v.dtd" as="pre" ?>
  <pre:element/>
  
Oh humm.  Thank you all for your feedback!

;) Clark



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/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe 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)