• To: xml-dev <xml-dev@l...>
• Subject: Re: ID-ness in XML
• From: "Simon St.Laurent" <simonstl@s...>
• Date: 11 Nov 2001 14:05:02 -0500
• In-reply-to: <E162yaE-0007zg-00@m...>
• References: <E162yaE-0007zg-00@m...>

On Sun, 2001-11-11 at 12:39, John Cowan wrote:
> Simon St.Laurent scripsit:
> 
> > I'd make xml:id something that can be used with or without a DTD.  If
> > used in a DTD, it would have to be declared of type ID and would have to
> > be the only ID on any given attribute.  That seems to avoid a lot of
> > pain to me.
> 
> Fair enough, but what does this "have to" mean?  On pain of invalidity?

Pain of ambiguity, I'd call it.  If you're crazy enough to use xml:id
AND another attribute of type ID, there will be pain down the line, sure
enough.

As Namespaces in XML didn't modify XML 1.0's rules for validity,
introducing a different class of constraints instead, I'd say this
probably should go the same route.

-- 
Simon St.Laurent
"Every day, in every way, I'm getting better and better." - Emile Coue

• References:
  • Re: ID-ness in XML
    • From: John Cowan <cowan@m...>

• Prev by Date: Re: ID-ness in XML
• Next by Date: Re: So maybe ID isn't a problem after all.
• Previous by thread: Re: ID-ness in XML
• Next by thread: Re: ID-ness in XML
• Index(es):
  • Date
  • Thread