[Home] [By Thread] [By Date] [Recent Entries]

  • From: Philip Nye <philipnye@f...>
  • To: "XML Developers' list" <xml-dev@x...>
  • Date: Wed, 08 Mar 2000 12:08:02 +0000

Toby Speight wrote:

> "WFC: Unique Att Spec" on production 40 forces uniqueness.  There's
> no such constraint on orderedness, though you may be unable to use
> standard parsers if you want to preserve order.

Thanks - I missed that one. This gives;

              Elements            Attributes
Ordered?      schema dependent    Probably not
Unique?       schema dependent    Yes
structured?   schema dependent    No

Which still shows that anything attributes can do elements can do too -
but not vice versa.

Henry's suggestion used a leading "." to flag an element to be
interpreted as an attribute but this leads straight to the question of
why use real attributes at all when the schema can be designed to do
whatever you want with elements?

If a particular schema language prevents you from doing all these things
with elements then because the three basic constraints are orthogonal,
there will always be problems. I think Henry's post illustrated one such
problem coming to the surface.


Philip

***************************************************************************
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/
***************************************************************************

Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member