RE: Classification and discovery

[David RR Webber <Gnosis_@compuserve.com>]

Message text written by Bob Haugen
>
>Bob,

>The GUIDE classification layer is designed to solve all this.

David, I am sorry to be sharp, but the BP and CC teams 
have been working on this stuff for many months.  
GUIDE was developed by you and I don't know who all else 
outside of ebXML (as far as I can tell) and then dropped in our laps 
just before Tokyo as if it was the answer to every aspect of
ebXML and all our hopes and dreams.

>>>>>>>>>>>>>>
   Bob - that's not accurate - GUIDE was developed
   prior to SJ - and was presented at SJ and formally
   then submitted to the ebXML Executive so that 
   the whole work is available for use in ebXML, just
   as IBM did with TPA.

  Just like TPA also - its was realized that there is 
  broad applicability to the concepts and approaches
  in GUIDE, not just to BP, but also CC, RegRep, TRP
  and so on.   The GUIDE details are there as a basis
  for ebXML to develop, refine, enhance, improve
  and incorporate.  There should be an ebXML GUIDE;
  RosettaNet has PIP's and they appear to be moving 
  toward a GUIDE-like mechanism beyond V2 of their
  work.  We can all align around a single approach if
  we are smart here.
<

Has any working group in ebXML adopted GUIDE?
Are you proposing that BP adopt it in place of everything
the group has worked on?
>>>>>>>>>>>>>>
 Bob - the whole idea is that GUIDE should provide an
 integration layer between BP, CC, RegRep, and TPA.
 The existing BP work should fold neatly into this model
 and I'm looking forward to doing just that - so that BP,
 and whoever else feel that they now have BP-GUIDE, et
 al that provides the XML implementation of the conceptual
 approach.  

 What I'm looking for is THAT detail on the BP
 structural classification expressed as XML.   My assumption
 is that the work EDIFECTS is rumoured to be doing is that.
 If anyone has that XML - please let me know.  If you do not
 have the XML - but have the information model done - great!
 We can create the XML from that.  As Nelson said 'I see no ships'.
<

>I hope, I really really hope, that BP has more than your thumb sketch 
>here to provide in the way of XML classification structures to 
>drive this process.  

Good grief, was that not apparent from my introduction?
Where have you been?
>>>>>>>>>>>>>>
  Bob - I'm looking for specific XML structures at this point.
  What you provided was an interesting written use case,
  but I have to say one that was a classic 
  programmer-centric stance of 'it works this way because
  because that's the technology'.   You will always find I
  will come at you the otherway around.  As an enduser 
  what do I want here?  I want the machine to be as simple
  and intelligent to use as possible.  The interaction path
  you described was, well convoluted.  Now that IS a 
 facet of the fact that we do NOT have this all cogently 
 integrated yet - but we have to design smart here.

 OK - you asked for criticism and I'm known for not 
 holding back in that department ; -)

 What is the good news?  Actually I included a section 
 in the RegRep draft for Tutorial and Test Drive, and 
 that is exactly where the dialogue you provided fits in.
 I definately want help in providing that and we'll wait
 to see what Farrukh as editor needs there (part of this
 is also the Tokyo PoC use case).

Hopefully this is all helping to clarify the needs.

Thanks, DW.