Hi,

At 08:49 PM 1/31/2010, Stuart wrote:
>Birnbaum, David J wrote:
>>Dear TEI-L,
>>
>>>There is a useful commentary on this in David J. Birnbaum, "In 
>>>Defense of Invalid SGML", ACH/ALLC
>>>1997, Queen's University, Kingston, ON. 
>>>http://xml.coverpages.org/birnbaumACH97.html
>>An updated version of that report is available on-line at:
>>
>>http://clover.slavic.pitt.edu/~djb/sgml/invalid.html
>
>At the NZETC, we solve this by resorting to schematron, which issues 
>warnings rather than errors. A significant minority of our texts 
>have schematron errors because the schematron reflects how we'd like 
>the world to work, rather than how it does.
>
>Yes, I understand that this is just moving the problem from DTD to 
>schematron, but almost all modern XML tools provide little or not 
>XML support once you venture off the DTD-beaten track.

In my experience, "just moving the problem from DTD to Schematron" is 
more significant and more helpful than it might appear at first.

This is because many (most) systems assume their DTDs will be 
gatekeepers -- "invalid documents shall not pass" -- which is 
entirely inappropriate to the kind of interesting and important 
irregularity that David wrote about.

This also goes to the heart of the requirements clash between 
prospective markup, i.e. XML encoding done for the sake of fitting 
data to a particular application or family of applications -- which 
typically require at least some forms of regularity and 
predictability -- and retrospective markup, which aims to *describe* 
documents irrespective of concerns with how regular they are, and 
which may be as interested in irregularities as in regularities. It's 
not that prospective and retrospective aims cannot be reconciled: 
they can. But to do so requires a fair amount of art.

To that end, placing "hard" validation requirements in a 
grammar-based schema such as a DTD or RNG schema -- but modelling 
loosely -- while relegating "soft", heuristic validation (which can 
potentially be much tighter, since to fail a test is not considered 
ipso facto an error) to a rules-based approach such as a Schematron 
or XSLT-based tag checker, is a very workable approach.

I wrote about all this back in my paper of 2001, "Beyond the 
'Descriptive vs Procedural' Distinction" --

http://piez.org/wendell/papers/beyonddistinction.pdf

Cheers,
Wendell



======================================================================
Wendell Piez                            mailto:[log in to unmask]
Mulberry Technologies, Inc.                http://www.mulberrytech.com
17 West Jefferson Street                    Direct Phone: 301/315-9635
Suite 207                                          Phone: 301/315-9631
Rockville, MD  20850                                 Fax: 301/315-8285
----------------------------------------------------------------------
   Mulberry Technologies: A Consultancy Specializing in SGML and XML
======================================================================