Re: [ubl-ndrsc] Proposed Modification to [ELN4] - ASBIE Naming

[Tim McGrath <tmcgrath@portcomm.com.au>]

looks like we are both in this together kid ;-)

i agree that we get distracted by the conceptual (SNM) view and the 
implementation (XSD) view.  but i dont think we are that far apart. it 
is not that we cannot nest/contain ABIE inside ABIEs - its that we dont 
need to.  it is the hierarchical nature of XML documents that means 
ASBIEs are implemented by nesting.

don't the XML instances look like....

<element> (may be ABIE)
    <element>  (may be BBIE)
    <element> (may be ABIE with its definition reflecting the context of 
its use - ie its ASBIE definition)
        <element>  (may be BBIE)
        <element> (may be another ABIE with its definition reflecting 
the context of its use - ie its ASBIE definition)

isn't the mapping the NDR are trying to do like this...?
ABIE = XSD type
BBIE = XSD type
ASBIE = reference to XSD Type (of ABIE)

from the example above it should also be clear that in any given 
document schemas, all the nesting will be ASBIEs - with the exception of 
the root ABIE.

as i said earlier, ASBIE are not elements/objects in the way ABIE and 
BBIEs are.  they are more abstract concepts and in XML terms we can only 
use documentation to show their existence.

Grimley Michael J NPRI wrote:

>Tim,
>
>BTW: Personally, I *may* be leaning towards your method of, where appropriate, simply having the element that represents the ASBIE reference the ABIE that it is associated to.
>
>I have a theory. I think part of the problem is that we refer to the elements themselves as ABIEs, ASBIEs, etc. when, in fact, they are not. They are the XML mapping/representation of the BIEs in the syntax neutral model. Because ABIEs cannot contain other ABIEs in the SNM, there is resistance to letting an element representing an ABIE contain another element representing an ABIE in our XML binding, which is what you are doing.
>
>I believe that by referencing an element, rather than declaring another element of the same type, you are placing that referenced element (documentation and all) into the content model of the containing element. Therefore, I'm not sure it is correct to reference an element whose documentation indicates it is an ABIE and call it (in its own documentation) an ASBIE. You are mapping an ASBIE in the SNM to an element that represents an ABIE. Again, I'm not sure this is totally wrong, it's just not what was intended by the rule.
>
>Does any of this make sense? Or am I totally off the wall? (or both?)
>
>MikeG
>
>  
>

-- 
regards
tim mcgrath
phone: +618 93352228  
postal: po box 1289   fremantle    western australia 6160