[
Lists Home |
Date Index |
Thread Index
]
On Jan 16, 2004, at 5:17 PM, jcowan@reutershealth.com wrote:
>> Given the desperate need to ensure that documents that
>> describe potentially high-priced financial instruments are correct in
>> their content, why doesn't it make more sense for you to kick back the
>> badly formed documents to their source and ask for clean versions?
>
> One reason is the existence of a settlement process. It's cheaper,
> quite
> often, to assume all is well, watch for exceptions further down, and
> correct
> them by hand.
A recent nuke in the Atomic war :-) over this issue in the Atom
community linked to the W3C SGML WG thread where this issue was
apparently aired for the first time back in early 1997.
http://lists.w3.org/Archives/Public/w3c-sgml-wg/1997Apr/0164.html began
it all, apparently.
One post I found particularly intriguing in hindsight was from Paul
Prescod
http://lists.w3.org/Archives/Public/w3c-sgml-wg/1997May/0074.html
"Browsers do not just need a well-formed
XML document. They need a well-formed XML document with a stylesheet
in a known location that is syntactically correct and *semantically
correct* (actually applies reasonable styles to the elements so that
the document can be read). They need valid hyperlinks to valid targets
.... There is still so much room for a document author to screw up that
well-formedness is a very minor step down the path. The idea that
well-formedness-or-die will create a "culture of quality" on the Web
is totally bogus. People will become extremely anal about their
well-formedness and transfer their laziness to some other part of the
system.
I' m not at all sure that the 2004 edition of Paul Prescod would agree,
but the idea that well-formedness is a particularly useful indication
that a document is "correct in content" seems a bit, well "bogus".
There's a lot to more "validity" in the real-world sense of the word
than well-formedness or even validity against an XML schema, even in
something as simple as a typical Web document let alone a business
document with legal and financial ramifications. Most likely, some
"settlement process" is needed for a wide range of errors, and
well-formedness is just the most easily detected.
I do think we in the XML world need to be careful not to overstate the
benefits of well-formedness even if we do insist that it is intrinsic
to the formal definition of what "XML" means. One could of course
argue that it is an indicator of non-"laziness". Is there much
evidence to support this? Does a working knowledge of the corner cases
in the XML spec (or the ability to choose software that handles them
properly) really correlate with overall document quality/validity, in
people's experience?
"
|
