ubl — archive

[Tim McGrath]

 MHonArc v2.5.0b2 -->


















ubl message






[Date Prev]
 | [Thread Prev]
 | [Thread Next]
 | [Date Next]

--

[Date Index]
 | [Thread Index]
 | [List Home]








Subject: Re: Code Lists (for Atlantic call if possible)     was Re: [ubl]NDR Codelist Example




From: Tim McGrath <tmcgrath@portcomm.com.au>
To: ubl@lists.oasis-open.org
Date: Wed, 07 Jun 2006 22:48:59 +0800










having seen Ken and Syliva's comemnts I think we are in  agreement. 
just a confusion between ATG defined codes and no-ATG defined codes.  I
have no issue with ATG defined codes - they are out of our control.  we
should be getting those from ATG (or the CEFACT ICG group). therefore
our NDRs have nothing to say about them.


my comments only relate to UBL defined codes - many of which are new to
UBL 2.0.


Tim McGrath wrote:

  
  
unfortunately i cannot hold out for the Atlantic call tonight but i do
think we need to get this straight.  as we try to finalize things,
starting to put the pieces
together, we need to understand how all the threads connect.  This
means the NDR, schemas and the code list approach have to support
each other.


i agree with Ken, i thought we had made a decision to use external
/second-phase validation (ie. genericode) for enumerations and not
mandate any methodology (e.g. W3C Schema, Schematron or XSLT).  if UBL
2.0 has the UBL1.0 code list schemas complete with enumerations
then i dont see how we can also have a genericode offering as well.  if
GEFEG users feel the need to continue with the 1.0 methodology then i
am not sure this is a problem. they can achieve this with a
supplementary, second-phase validation schema.


perhaps to move ahead with this issue we need to build an end-to-end
example to show what we think the mechanisms should be. i know Ken did
this but now was can use real UBL 2.0 examples. I suggest we take one
UBL 2.0 code list and show:

a. the spreadsheet entries

b. the UBL 2.0 schema(s) view

c. the genericode file for validations

- and that's it. 


i will volunteer for (a) and (c) but may need help with (b).  once we
have this we can have a more informed discussion about the implications
of all this.  then we write the NDRs that support this approach.


Sylvia Webb wrote:

  

    
The current NDR does support the new codelist methodology. This is not the
question.

As I stated in Ottawa and in February, GEFEG received substantial feedback
that UBL users need to continue to use the UBL 1.0 codelist methodology,
regardless of any opinions about that methodology. This means that as an
option, the NDR need to continue to include existing portions of Section 6.

Possibly Section 6 needs additional clarification, but, this should not mean
deleting the previous rules.

Sylvia