On Thursday, August 02, 2001 10:01 PM
Syd Bauman wrote:

> (The crazy analysis of SIC w/CDATA, CORR w/IDREF
> I don't buy; note that both ABBR and EXPAN use IDREF,
> but both ORIG and REG use CDATA.)

Crazy it certainly was (paranoid was my own word for it) but there are
some crazy folks out there (with the sheer difficulty of making sense of
the Guidelines when your boss is yelling deadlines at you being a
contributory factor). Faced with what are in this instance arbitrary
inconsistencies in a document which otherwise shows every indication of
immense care and reflection, the normal scholarly response is a kind of
hermeneutic equivalent of Pascal's wager. And I think what you say
actually illustrates precisely how, again by pure chance, this
particular constellation creates a hemeneutic mirage. That both ABBR and
EXPAN use IDREF, but both ORIG and REG use CDATA makes the
likely true explanation plausible enough, viz that different people
probably wrote/tested/reviewed on different days, since the pairs
that "naturally" belong together have matching types. But not
SIC and CORR. They obviously must have been finalised by the
same person(s) at the same time, and yet the types differ. And so our
seeker of enlightement, already pondering why we need both
SIC *and* CORR, thinks a light has dawned...

Michael
---------------------------------------------------------
Michael Beddow   http://www.mbeddow.net/
XML and the Humanities page:  http://xml.lexilog.org.uk/
---------------------------------------------------------