XML-Data: advantages over DTD syntax?

[Rick Jelliffe]

> From: Henry S. Thompson <ht at cogsci.ed.ac.uk>

> In the area of document grammar specification, XML-Data offers no
> functionality which cannot be duplicated by extensive use of parameter
> entities, or at worst tedious by-hand expansion.  But it provides that
> functionality in a transparent (in some cases MUCH more transparent)
> way, and in my view that makes it worth bothering.

I call as exhibit #1 FrameMaker's EDD (element description definition???)
format.   The developers of XML-data should look hard at it, and the
lessons to be drawn from it.  It seems to have been conceived as a
better SGML than SGML  (Frame also had an additional requirement to 
embed structure into their interchange format too). 

It is more friendly/verbose than SGML's declarations, provides 
slightly more expressivity, slightly better attribute types,
and includes style specification for element-types-in-context.
It also has a cool syntax-directed editor.

It is certainly far more developed than XML-data (in that it
has been developing and in use for a few years now), though
it does not use XML elements for declarations nor use inheritance
mechanisms.  I have watched EDD with interest, and the first
comment to make is that they have had to match SGML's primitive
capabilities, even in things that did not seem requirements
to them at first.  XML-data would have to do the same, I'd
expect, unless it offers significant benefits in some new area.

EDD has become a very large product, but I found that using
it (in the recent FrameMaker+SGMLs) was a little tedious, 
in that it did not offer such enormous advantages to make 
warranted the duplication of SGML declaration in its own syntax.

For exhibit #2, I call the Pinnacles or DOCBOOK DTDs, expressed
in XML-data.  Can someone whip it up, and we can get a much better
feel for how readable it is as a declaration syntax for a nice
juicy DTD?  The number of derived element types will probably be
much fewer than the number of base element types, surely. Without
exhibit #2, I really don't feel comfortable making claims that
XML-data is verbose (or reading claims that is is more transparent!)
 

Rick Jelliffe
 



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)