Re: [ubl] Minutes of Atlantic UBL TC call 29 June 2005

["Stephen Green" <stephen_green@seventhproject.co.uk>]

> MavisC/MikeG: The NDR editors discussed BryanR's request for an
>   ANY area.  We don't have an objection to this, but think that a
>   better solution might be to use the UBL documents as specified
>   in our schemas and then embed them in a wrapper containing the
>   data needed for local requirements.
>
>   JonB: Or better yet, in another part of the message (for
>   example, another MIME part in an ebXML message).
>
>   StephenG: Or in the CEFACT document header.
>
>   AGREED to record this and continue the discussion next meeting.
>   BryanR is invited to join the NDR editors in their conference
>   call immediately preceding the Atlantic meeting 13 July.


>    Continue the discussion of ANY and the other requests from
>    BryanR, to wit:
> 
>       (1) A reconsideration of our prohibition of XSD ANY, because
>       there are regional laws requiring the inclusion of specific
>       information, and we need an extensible content area to
>       handle this; (2) Restrictions on strings in UBL content to
>       ensure that the content consists of more than white space,
>       for example through length or minlength facets; (3)
>       Reconsideration of our prohibition of appinfo, because there
>       are many cases where one element is conditional on another;
>       this would give Scehamatron (for example) the data it needs
>       to do conditional/contextual validation.
> 

I just sent out a message on ubl-dev related to this (coincidentally
since I had a similar matter I needed to consider).
http://www.oasis-open.org/archives/ubl-dev/200507/msg00018.html

My matter concerned not adding data to UBL documents but
adding UBl documents to other documents (in my case an
XBRL-GL document *after* the sending and receiving of
the UBL message). In this case it seems appropriate to use
a header such as SBDH because it isn't a matter of adding
another XMl document to UBL, but for some the latter may
be an acceptable proceedure.

I suggested that there could be an xsd:type for XML and since
it was pointed out to me (thanks Chee-Kai!) that xsd:any is just
that. 

So I'd suggest we add (or better still ATG2 add) a datatype
or set of datatypes like 'XMLType' (or 'StructuredDataType', say)
1) based on xsd:any or something like it (xinclude?)
and/or another
2) based on xlink or something like it
which would allow, completely at the modelers' discretion, the
1) inclusion and/or 2) referencing of structured (XML) data
in an element where it is appropriate to link, add or in some other
way associate XML from another schema in a UBL document.
The metadata attributes (supplementary components) should include
the schema information of the linked, included or associated
XML.


I'd also suggest we might try picking up on old liaisons with
XBRL to see how best practise guidelines might be developed
for the special case of linking UBL documents to XBRL 
(especially, perhaps, GL) documents.


Further considerations of how the pitfalls can be avoided
would be just as much warranted as with the addition of
the cbc:Note based on xsd:string but this was really a
modeling issue rather than an NDR one. I suggest this matter
is similar and by providing the mechanism recommendation
to the modelers the pitfall considerations can be defered to the
same too.

All the best

Stephen Green