• To: <xml-dev@l...>
• Subject: Re: So maybe ID isn't a problem after all.
• From: Sean McGrath <sean.mcgrath@p...>
• Date: Mon, 12 Nov 2001 09:17:15 +0000
• In-reply-to: <00d801c16a06$a3fb7bc0$0301a8c0@n...>
• References: <9A4FC925410C024792B85198DF1E97E401C9B2EC@u...>

At 11:42 10/11/2001 -0500, Jonathan Borden wrote:
>  Mike,
>
> >
> > > -----Original Message-----
> > > From: Jonathan Borden [mailto:jborden@m...]
> > > Sent: Saturday, November 10, 2001 11:00 AM
> > > To: Champion, Mike; xml-dev@l...
> > > Subject: Re:  So maybe ID isn't a problem after all.
> > >
> >
> > > so what am I missing?
> >
> > Maybe nothing, except my wistful hope that we really can do this kind of
> > thing with WF XML syntax  rather than DTD syntax someday...
>
>I know what you are getting at, but want to remind people that "well-formed
>XML" _includes_ the internal DTD subset and hence all of this can be done
>in-band.

Yes it can be expressed in WF XML, but internal subset stuff disappears
to Tumbolia  in XML processing chain/workflows.

Four alternatives I see:
         Don't use internal subsets

         Use internal subsets and live with the fact they they don't 
survive processing pipelines

         Make the internal subsets round-trip by using a sufficiently rich 
and complex
         infoset.

         Write monolithic XML processing programs


regards,
Sean

• References:
  • RE: So maybe ID isn't a problem after all.
    • From: "Champion, Mike" <Mike.Champion@S...>
  • Re: So maybe ID isn't a problem after all.
    • From: "Jonathan Borden" <jborden@m...>

• Prev by Date: SV: Caught napping!
• Next by Date: Re: So maybe ID isn't a problem after all.
• Previous by thread: Re: So maybe ID isn't a problem after all.
• Next by thread: RE: So maybe ID isn't a problem after all.
• Index(es):
  • Date
  • Thread