[EMAIL PROTECTED] wrote:
>In a message dated 2/12/2004 10:39:42 AM GMT Standard Time,
>[EMAIL PROTECTED] writes:
>
>
>>In SGML and XML, a document is composed of two sequential parts,
>>the prolog and the instance. You can see this in an HTML example:
>>
>>1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
>>2 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
>>3
>>4 <head>
>>5 <title>The Symbol Grounding Problem</title>
>>6 </head>
>>7
>>8
>>9
>>
>>In this example, the prolog is lines 1-2, the instance begins on
>>line 3.
>
>
>
>Murray,
>
>That is incorrect. The prolog includes lines 1 - 3.
No. The document instance is the element, which begins
on line 3.
Murray, please review the relevant parts of the XML Recommendation which defines the prolog and you will, I think, see that you are wrong.
The only prolog content in my example was the DOCTYPE
declaration.
No, that's not correct either. The whitespace in line 3 is part of the prolog. Check the XML Recommendation if you don't believe me.
In a subsequent email message, Vladimir brought up
the XML declaration:
Vladimir Cvetkovic (AT/LMI) wrote:
>Here is the first line of the document:
><?xml version="1.0" encoding="UTF-8"?>
XML modifies SGML by optionally adding this to the prolog. So it
would be line 0 in my example.
>The prolog includes the DOCTYPE declaration, the external
>>subset (called the DTD), and the internal subset (which you seldom
>>see but it's legal).
>
>That too is incorrect, in my view. The prolog may optionally include a
>pointer to the external subset but the external subset is not part of the prolog,
>although it is part of the DTD.
I'm sorry, but I've been doing SGML and XML for well over a decade,
have been an author/editor of a number of industry DTDs (including
the modularization of XHTML and the XTM DTD), but rather than spout
my CV at you
Didn't you just do that? :)
, please read the relevant passages from either ISO 8879,
Goldfarb, or any book on SGML.
Are you suggesting that such an impressive CV is a sure sign of infallability? :)
Neither ISO 8879 (as modified) nor Goldfarb's SGML Handbook define XML authoritatively. In any case I am not convinced that Goldfarb states about SGML anything different to the point I was making.
However, we are talking about XML here, not SGML.
>BTW I did check If you want to get picky about language,
the declaration subset contains the DTD.
Can you point me to where you believe the XML Rec states that?
It seems to that the DOCTYPE declaration can contain the internal subset and point to the external subset.
Andrew Watt
That can be as an external
subset (the most common, since it's an external reference to a document
called a DTD) and/or an optional internal subset. But all parts of the
declaration subset, including the external and internal subsets, are
part of the prolog.
