[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [ubl-fpsc] Code lists as a repository of code list supplemental information
>>> >><xs:enumeration value="USD"> >>> >> <xs:annotation> >>> >> <xs:documentation> >>> >> <xhtml:div class="Code_List._Expansion._String"> >>> >> <xhtml:p>US Dollar</xhtml:p> >>> >> </xhtml:div> >>> >> <xhtml:div class="Code_List._Expansion._Symbol"> >>> >> <xhtml:p>$</xhtml:p> >>> >> </xhtml:div> >>> >> </xs:documentation> >>> >> </xs:annotation> >>> >></xs:enumeration> >> >>The latter ... that any product or technology might read the XSD >>expressions and distill the two transliterations: >> >> "USD" -> "US Dollar" >> "USD" -> "$" >> >>.. and I suppose for input and validation purposes: >> >> "US Dollar" -> "USD" In that case, I thought of the following: (1). xhtml wouldn't be the "right" way to store them as XHTML is intended for presentational data, less so for informational data. (2). In LC, we're converting the previous attribute-based schema metadata to element-based storage, in the standardized form of : <xsd:annotation> <xsd:documentation> <ccts:Component> <ccts:CategoryCode>ABIE</ccts:CategoryCode> <ccts:ObjectClass>Address</ccts:ObjectClass> ... </ccts:Component> </xsd:documentation> </xsd:annotation> Might FP consider proposing useful element structure to store as part of the ccts space? I'm imagining something like: <xsd:annotation> <xsd:documentation> <ccts:Component> ... </ccts:Component> <ccts:Presentation> <!-- FPSC proposed structure --> </ccts:Presentation> </xsd:documentation> </xsd:annotation> If we use this structure, I think it'll standardize storage of all the various aspects of metadata pertaining to the schema. Would this be more appropriate? (3). And this one's for you, Lisa. Given that Ken's use of the data is for processing purposes rather than to provide in-line documentation, and that we could hardly expect not to have metadata meant for machines when XML's advantage is that it is machine processable, I was wondering why there was an earlier decision by NDR to rule out categorically any use of <xsd:appinfo>. I also find those <ccts:Component> metadata appears more appropriate for processing purposes than as documentation of the schema, since many downstream tranformations could rely on the ObjectClass, RepresentationTerm etc properties to drive their transformation behaviors. I'd be happier if we use <xsd:appinfo> than <xsd:annotation> for such machine-oriented purposes. Another battle-front for NDR? Best Regards, Chin Chee-Kai SoftML Tel: +65-6820-2979 Fax: +65-6743-7875 Email: cheekai@SoftML.Net http://SoftML.Net/
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]