Adopting EDIFIX for UBL 1.1

[jon.bosak@sun.com]

Hello UBL TC,

We've recently been discussing the proposal from GEFEG to use
EDIFIX (EF) as our primary data model management tool for UBL 1.1.

   http://lists.oasis-open.org/archives/ubl/200410/msg00035.html
   http://lists.oasis-open.org/archives/ubl/200410/msg00036.html
   http://lists.oasis-open.org/archives/ubl/200410/msg00041.html
   http://lists.oasis-open.org/archives/ubl/200410/msg00049.html

We seem to be in general agreement that we should accept GEFEG's
proposal to use EF for 1.1, so I'd like to go ahead and formally
propose this to the TC.

I'd like to emphasize the following points, most of which have
already been covered somewhere in the discussions referenced above
but need to be made explicit.

 - Before we can begin to use EF, we will need to establish that
   the spreadsheet format we have been using can be imported into
   EF losslessly and exported from EF losslessly.  I intend to
   schedule this work as part of the meeting in Santa Clara.

 - The ability to import spreadsheets into the EF data model is an
   absolute requirement because we need to support outside groups
   submitting UBL 1.1 adds in that format.  Whether this is how
   the team(s) we establish to make such changes actually
   choose(s) to operate should be up to the people involved; it's
   possible that a given team will prefer to instruct the person
   we put in charge of EF input to make some changes directly and
   rely upon the generated spreadsheet as documentation.

 - Similarly, the ability to generate spreadsheets from our model
   will continue to be an absolute requirement, not just for
   purposes of documentation but so that external groups can
   review the model in that form and use tools such as UBLish to
   check the model against the schemas.

 - The fact that our data model gets converted into a different
   format upon entering EF does not make it any less our data
   model.  In other words, it does not stop being OASIS IP under
   the control of this Technical Committee just because its
   internal representation has changed.  GEFEG has said that they
   understand this.

Assuming that we're clear on all of this, I hereby propose to the
TC that we adopt the proposals outlined in the reports referenced
above.  If you have no objection to adoption of this proposal, do
nothing.  If you do object to adoption of this proposal and are a
voting member of the TC, please register that objection on this
mail list no later than COB Tuesday 26 October 2004.

Jon Bosak
Chair, OASIS UBL TC