RE: [xml-dev] Re: URIs, concrete (was Re: [xml-dev] Un-ask the question)

From
Simon St.Laurent <>
To
XML DEV <>
Date
2002-07-31T22:21:48Z
ID
<r01050300-1015-E7E4D196A4D311D693BB0003937A08C2@[192.168.124.21]>
Thread
RE: [xml-dev] Re: URIs, concrete (was Re: [xml-dev] Un-ask the question)
> Simon, The feature isn't useless. Most people agree that attributes
> are meant for annotating elements. The Namespaces in XML
> recommendation allows authors to choose between context sensitive
> attribute values E.g.
> 
> <myns:Parent name="must be a number between 1 - 100"><myns:Child
> name="can be any string"/></myns:Parent>
> 
> or creating global attributes that represent the same idea regardless
> of what element they appear on
> 
> <myns:Parent myns:id="must be unique in document"><myns:Child
> myns:id="must be unique in document"><></myns:Parent>

I'm sorry, Dare, but in my world that's utterly useless, not to mention
confusing and complicating.  If you can find a genuinely good reason for
this:

<rdf:Description rdf:about="http://www.w3.org"
about="http://www.w3.org/not/really/I'm/just/kidding/">

please let me know.  Using global attributes to mean something different
in a local context seems completely bizarre and desperately pointless to
me.  It's bad practice.


-- 
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com