Re: Intentions of XMP

_____________Original message ____________
Subject:	Re: Intentions of XMP
Sender:	ext Brian McBride <bwm@hplb.hpl.hp.com>
Date:		Thu, 26 Sep 2002 21:34:31 +0300

At 18:24 26/09/2002 +0300, Patrick Stickler wrote:
>
>         Perhaps a clearer, more mnemonic way to ask this question
>         would be, do the literal values of the following two properties
>         mean the same thing to XMP applications? Would they be
>         considered to carry equivalent semantics in both cases?
>
>         <xmp:CreateDate>2002-09-25T11:36:07Z</xmp:CreateDate>
>         <dc:title>2002-09-25T11:36:07Z</dc:title>

Oh dear, that's not the same question at all.  If we were going to ask 
Adobe, it would have been useful to agree the question first.

	Oh come, now Brian. It precisely matches the structure
	and semantics of the generic entailment. Given two
	occurrences of the same inline literal with two different
	properties, can we conclude that the interpretation of the
	literal in both cases is the same.

	If one presumes that the inline literals denote strings, then
	the answer is yes. If one presumes that the literals denote
	values, then the answer depends on the context of
	interpretation, and in my example to Adobe, that context
	leads to an answer of no.

	If two properties were chosen which happened to assert
	the same datatype, then the answer would be yes, and we
	wouldn't be any the wiser about the presumptions leading
	to that answer. This is why this particular entailment
	question is problematic. Because one may arrive at a
	yes answer based both on tidy and untiidy presumptions,
	if the properties do not force consideration of the context
	of interpretation.

	I think that the comments from Adobe are more than
	sufficiently clear, that XMP employs a value based
	interpretation if inline literals in precisely the same
	fashion as CC/PP.

	Patrick

Received on Thursday, 26 September 2002 14:50:44 UTC