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

RE: linking, 80/20

• To: 'John Cowan' <jcowan@r...>, elharo@m...
• Subject: RE: linking, 80/20
• From: "Bullard, Claude L (Len)" <clbullar@i...>
• Date: Tue, 20 Aug 2002 08:11:50 -0500
• Cc: xml-dev@l...

Actually, I was talking about the level of elements 
and attributes.  The XML document level doesn't mean 
very much without type.  It's just tag soup with 
a coincidental outer tag.  In practice, I don't 
believe it either.   

Namespaces are something to be wary of, not to 
be built over.

len


From: John Cowan [mailto:jcowan@r...]

Elliotte Rusty Harold scripsit:

> >Naah.  Namespaces are designed to prevent name collisions and that's it.
> >Anything else is all in your head.
> 
> That's the common refrain, and it may be true historically, but in 
> practice I don't believe it. The primary use I see for namespaces is 
> to quickly and easily recognize elements from particular 
> vocabularies, even in the absence of local name conflicts. Resolving 
> name conflicts is actually quite rare.

I agree with that at the level of element types and attribute names.
I don't agree at the level of documents, which is the level Len was
talking about (and I suspect you don't either): there is no such thing
(except by accident) as "a document in a namespace".  Multi-namespaceness
is the true nature of XML documents.

• Prev by Date: Re: the infoset is two infosets (or even three?) [was: Re: linking, 80/20]
• Next by Date: Re: the infoset is two infosets (or even three?) [was:
• Previous by thread: Re: linking, 80/20
• Next by thread: RE: linking, 80/20
• Index(es):
  • Date
  • Thread