• From: Liam Quin <liam@w...>
• To: Elliotte Rusty Harold <elharo@i...>
• Date: Sun, 23 Aug 2009 19:14:03 -0400

On Fri, Aug 21, 2009 at 09:21:50PM -0700, Elliotte Rusty Harold wrote:
> On Fri, Aug 21, 2009 at 7:46 PM, Liam Quin<liam@w...> wrote:
[...]
> We don't have to speak in infoset language here. The critical points are:
> 
> 1. All conformant parsers MUST process the internal DTD subset.
Agreed.

> 2. The DTD, including the internal DTD subset, can define new entity
> references.
Agreed.

> If they're defined in the internal DTD subset, the parser
> must expand these.
It must at least make the expansion available.

> 3. Nothing except a DTD can define entity references.
Right. (although the ISO Schema people seem to have other
ideas, see the DSRL draft for an example)

> There are other details (default attribute values, malformed internal
> DTD subsets, external entity references) but those are what really
> make it impossible to just ignore DTDs or claim they're just one kind
> of schema language.

I'd say they are a kind of schema laguage that's given special status
by the XML specification.  Ignoring the internal subset isn't an
option, for sure.

Best,

Liam


-- 
Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/
http://www.holoweb.net/~liam/ * http://www.fromoldbooks.org/

• References:
  • Re: Wikipedia on XML
    • From: Elliotte Rusty Harold <elharo@i...>
  • RE: Wikipedia on XML
    • From: "Len Bullard" <cbullard@h...>
  • Re: Wikipedia on XML
    • From: Kurt Cagle <kurt.cagle@g...>
  • Re: Wikipedia on XML
    • From: Liam Quin <liam@w...>
  • Re: Wikipedia on XML
    • From: Elliotte Rusty Harold <elharo@i...>