[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: namespace prefixes within attribute values

• From: David Carlisle <davidc@n...>
• To: xml-dev@X...
• Date: Fri, 10 Mar 2000 11:24:28 GMT

Eve Maler wrote

  I can see using QNames attribute values when you're manipulating element
  types and attribute names, as is done in XML Schema and XSLT.  However,
  in XLink, QNames are used -- in the true Namespaces specification sense
  -- for values that have nothing to do with element types and attribute
  names.

The same is true of xslt. QNames are used as the names of most constructs
(variables, modes, etc) not just elements and attributes.

  A lot of people are hoping that one day they'll be able to point to some 
  kind of document (compound or not...whatever) that will serve as the 
  definitive source for information about a namespace.

I think it's probably safer keeping namespaces just to serve for
distinguishing things that would otherwise have the same name
(such as p in html and p in something else) and to look to
schema for information about things (although not about what constitutes
the set of names in a namespace).

  I want to define the name "popupFor10Seconds" in my namespace definition 
  document, have I now precluded the possibility of having a 
  <popupFor10Seconds>? an <x myco:popupFor10Seconds="..." />?

well if there were to be a `namespace definition syntax' (which I think
would be a bad idea) then it would already have to cope with elements
and attributes having the same name, so it could presumably have to
cope with arbitrary datatypes: xlink names, xslt variables, xslt modes, 
xslt template names, etc. All these things may have the same name in the
same namespace.

David

***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@x...&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************

• Follow-Ups:
  • W3C Rant
    • From: "Matthew Gertner" <matthew@p...>

• Prev by Date: XED was buggy from 2000-03-08 1721Z until 2000-03-10 1056Z
• Next by Date: Clarification (was Re: Gutenberg Project <longish>)
• Previous by thread: Re: Canonicalized Schemas
• Next by thread: W3C Rant
• Index(es):
  • Date
  • Thread