Re: [xsl] XML Schema/XSL conflict (can't differentiate unqualified locals)?

Subject: Re: [xsl] XML Schema/XSL conflict (can't differentiate unqualified locals)?
From: David Carlisle <davidc@xxxxxxxxx>
Date: Tue, 9 Oct 2001 11:29:16 +0100

   I think namespaces are intrinsically very confusing.

   (1) people don't understand that there is no "resource" that the URI
   identifies

Of course this is made rather worse by the schema WG's decision to make
the namespace URI effectively the default location of the schema.


   (2) people don't understand why the rules are different for elements and
   attributes

Rules is rules.

   (3) people don't understand whether or not prefixes are supposed to be
   significant


Of xourse XPath (and its followers) makes this more subtle than a
reading of the namespace spec would make appear. The namespace spec
really implies that it would be trivial to namespace normalise a
namespace conformant document to change prefixes etc. But using prefixes
in attribute values (and possibly, but not in XSLT) element content
makes this a lot trickier.

   (4) people don't understand the arbitrariness of some of the rules, e.g.
   that the default namespace can go out of scope for descendants but an
   explicit namespace can't

You can't unmap a prefix once bound, whereas you can unbind the null
prefix, but is this really very confusing, do people often need to
locally unbind the default namespace anyway>? (I'd say almost never
except for the rather bizarre use of unqualified local elements in
schema)


   (5) people don't understand the difference between namespace nodes and
   namespace declarations

True.

   (6) people don't understand the difference between namespace declarations
   and attributes

True.

   In my view, namespaces are a disastrously bad piece of design, almost as bad
   as DTDs. All these confusions derive directly from the decision to add a
   fundamental architectural feature to XML as a "bolt-on extra", which is
   almost always bad software engineering.

On the other hand it was politically very important in the beginning to
have SGML compatibility, and perhaps a architectually cleaner solution
would just have failed in the marketplace (but who knows:-)

David




_____________________________________________________________________
This message has been checked for all known viruses by Star Internet
delivered through the MessageLabs Virus Scanning Service. For further
information visit http://www.star.net.uk/stats.asp or alternatively call
Star Internet for details on the Virus Scanning Service.

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Current Thread