RE: [xml-dev] So maybe ID isn't a problem after all.

["Champion, Mike" <Mike.Champion@SoftwareAG-USA.com>]

> -----Original Message-----
> From: Gavin Thomas Nicol [mailto:gtn@rbii.com]
> Sent: Sunday, November 11, 2001 1:19 AM
> To: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] So maybe ID isn't a problem after all.
> 
> 
> People understood XML pretty well. 

Well, if you mean they understand elements, attributes, and text in a
default encoding, yes.  I think most people  originally grokked XML by
analogy with HTML -- "OK,it's like HTML, but a little bit more: all the tags
have to be balanced, you really do have to quote attribute values, and you
get to choose your own tagnames ..." If I'm right, they *expect* to be able
to do the other stuff you get for free in HTML, such as links.

I've been waffling on DTD internal subsets vs PIs vs namespaces as a way to
define ID-ness; I think Daniel Veillard's post finally catalyzed my opinion:

" If you put xml:id="foo" on an element then blah.xml#foo will point to it".
People who think of XML as HTML++ won't already understand DTDs, or PIs, or
a namespace that they have to declare for themselves, but if they had
xml:id, they could link to it almost exactly as they do in HTML today. 


> For a lot of the common folk it's hard 
> to understand  what they need and what they don't:, becasue they have to 
> learn everything in  order to decide...

Good point ... in XML writ large, you have to know something before you know
whether you need to know it :~)  The virtue of the xml:id solution is that
the people who think XML is HTML++ probably think it's there in XML already!