UBL Naming and Design Rules SC

 View Only

[ubl-ndrsc] Tag structure work

  • 1.  [ubl-ndrsc] Tag structure work

    Posted 03-13-2002 10:54
    I have attached a rough draft of Arofan's tag structure writeup (a 
    revision of Mark's original one) here.  Disclaimer: I believe both 
    Arofan and Mark feel this is not ready to be seen by the group.  But I 
    also think it's essential for us to work through our agreements to 
    date, and using this document as a tool for this is the best way I can 
    think of.  I also think that the document isn't bad as a rough draft!
    
    Our decisions at the last F2F were pretty clear and pretty voluminous:
    
       http://lists.oasis-open.org/archives/ubl-ndrsc/200201/msg00050.html
    
    We also made some relevant tag structure decisions in telecons:
    
       http://lists.oasis-open.org/archives/ubl-ndrsc/200202/msg00057.html
    
    However, we need to ensure that these decisions are all represented in 
    the writeup.  We also need to crystallize things even further, 
    particularly around the dictionary question.  Here are some notes and 
    questions to get us started in discussing this:
    
    - What are the kinds of things in our dictionary?
       . XSD complex types: Are these equivalent to object classes? to
         BIEs? Are they always ABIEs?
       . Global top-level elements, one per document type: What are these
         equivalent to in the CC universe?
       . Local unqualified elements per complex type: Are these equivalent
         to object properties?
       . (Some) fully qualified element paths: What are these equivalent to
         in the CC universe?
       . UBL representation terms and their mapping to ebXML RTs
       . XSD simple types: Are these equivalent to representation terms,
         or is there a many-to-one relationship? Are these BIEs? Are they
         always BBIEs?
    
    - What is the mechanism for building and maintaining the dictionary
       and the other related deliverables? (this is largely a TTSC matter,
       but I want to be sure we understand it)
       . Spreadsheets should be the source and have sufficient fields
       . Schema modules are generated
       . User-friendly, hyperlinked documentation is generated
       . Is UML generated? What mapping rules are there?
       . Is the current Library Content SC spreadsheet format missing
         anything?
    
    I hope people get this before the meeting starts...
    
    	Eve
    -- 
    Eve Maler                                    +1 781 442 3190
    Sun Microsystems XML Technology Center   eve.maler @ sun.com