- From: <lee@sq.com>
- Date: Thu, 7 Nov 96 13:53:29 EST
- To: W3C-SGML-WG@w3.org, paul@arbortext.com
> what SGML authoring tools
> would produce valid XML given document instances in your Shakespeare
> and Religious Works collections?
>
> Assuming neither Author/Editor nor Adept Editor (whose output, I
> believe, are very similar) are answers to the above question,
A/E isn't (I think) because
(1) it won't let you put a PI before the DOCTYPE line
under any circumstance right now
(I haven't tried opening such a file in A/E, the assumption being
that we will add XML support to A/E and HoTMetaL Pro)
(2) it doesn't handle whitespace in the same way (nor in the SGML way):
all whitespace is retained. This (it has been argued) is desirable
in an editor.
(3) (currently) it does not handle EMPTY elements as per XML.
There aren't any EMPTY elements in those DTDs; the version of the
Psalter that Harry Gaylord and Robin Cover did is a better example,
as they used EMPTY elements to represent the overlapping structures
in that document that SGML tools can't easily handle. (CONCUR would
have been the obvious choice, although the TEI has a better way, as
so often, using a linked list of id/idefs)
(4) the documents use a public identifier in the document to get at the
DTD, and not a SYSTEM identifier that's a URL.
No existing collection of SGML documents published with Panorama will
work unchanged in XML.
> what
> would be the list of modifications necessary to convert what they would
> produce given document instances in your Shakespeare and Religious
> Works collections into valid XML?
I'll let Jon answer this :-) but most of it follows from what I have listed.
Lee
Received on Thursday, 7 November 1996 13:53:46 UTC