ebBP 7/17/2006: UMM and Transaction Patterns Discussion

["Monica J. Martin" <Monica.Martin@Sun.COM>]

As we discussed in last week's meeting, herein are the question raised 
by Pim re: the transaction patterns, responses on ebXML-dev and my 
response recently. Thanks.

> vanderEijk: I read in the ebBP minutes that you were thinking of 
> responding to my
> question. In the mean time, this is what I got from Chritian Huemer.  
> Is this similar
> to what you were thinking of?

mm1: ...If you look at the request-confirm information that we provided 
in the pattern matrices in v2.0.3, the response that Christian provided 
was consistent with your discussion and what is in the specification 
(Many details exist in specification in Section 3.4.9)....

I also should pull out some of our historical discussions where we spent 
(literally) hours talking about the patterns, their use, and how to 
enable understanding and effective usage of the patterns. Perhaps we can 
discuss how these were envisioned by folks focused on their implementation.

Much of the discussions occurred in 2004 and early 2005 and was more so 
concentrated on Commercial Transaction (Business Transaction), 
Notification and Request Response.  When we concretized the patterns in 
the schema, that drove much discussion in order to understand what would 
be used and why.  I am uncertain about pre-editor context validation. 
There was no view or specifics about context in the UMM related to BT 
patterns until we added it in ebBP v2.0.x.  There was the notion of 
context related to information however.

Thanks.

> -----Original Message-----
> From: Christian Huemer [mailto:ch@mminf.univie.ac.at] Sent: 26 June 
> 2006 14:01
> To: Pim van der Eijk
> Subject: Re: FW: [ebxml-dev] UMM Business Transaction Patterns
>
> Dear Pim van der Eijk,
>
> I am sorry for the late response. However, I was only one day in the 
> office
> since you sent it. I was in Germany for the football championships, at a
> meeting in Mechelen, one day in office agreeing with Birgit that I will
> answer your mail, and last week at SWIFT in Brussels for the TMG meeting
> where we had no internet access at all.
>
> We already understood that hardly understand the definition for
> request/confirm, which was input from the RosettaNet Framework 
> developed by
> EDIFECS and contributed to CEFACT in 2000. In the beginning, I thought 
> that
> as a non-native speaker, I am the only one who does not understand it.
> However, over the time I recognized that even non-native speakers do not
> understand it. Thus, in the development of the new UMM version started 
> last
> year, we changed the definition. So since last year we do not use the old
> definition anymore. The new definitions were not invented from scretch -
> they already existed before as part of other UMM explenatory 
> documents. We
> just used them as definitions.
>
> Please find as an attachment the final working draft of the UMM 
> foundation
> module 1.0 which was agreed last week at our TMG meeting. This final 
> working draft is now in the phase of implementation verification
> before becoming a UN/CEFACT specification. It is not even out on the 
> web -
> remember we had no internet access last week . At http://www.untmg.org 
> you
> still find the second working draft, but it will be replaced by our web
> master very soon. You will find the definition of transaction on page 
> 69 of
> the final working draft.
>
> I hope this helps. I finally try to explain the old definition 
> "pre-editior
> context validation": The transaction refers to a previously established
> contract. This contract sets the context. The reply depends on this 
> context.
> Thus, the replyer (person, system) must relate to the contract before
> answering. At least this was always my interpretation. However, I hope 
> that the new definition is more useful for the user community.
>
> Best regards,
> Christian Huemer 
>
>> vanderEijk: Dear Professor Huemer,
>> Klaus Naujok suggested that I contact you on this.   I am working on a
>> project (Netherlands government) where we are looking to apply the
>> ebBP/BPSS/UMM business transaction patterns.   We ran into the 
>> problem that
>> we cannot find an explanation for the concept op "pre-editor context 
>> validation".  Can you provide me with a definition or explanation of 
>> this phrase.  What context is being validated, and what does 
>> "pre-editor" mean?
>
>> Hopefully you can help us with this, as it will make ebXML adoption 
>> easier.
>>
>> -----Original Message-----
>> From: Klaus-Dieter Naujok [mailto:klaus.naujok@comcast.net]
>> Sent: 10 June 2006 17:50
>> To: ebxml-dev@lists.ebxml.org
>> Subject: Re: [ebxml-dev] UMM Business Transaction Patterns   
>>
>>> vanderEijk: I am in fact looking for a (reference to a) definition 
>>> of "pre- editor context validation".     
>>
>> naujok: You may want to contact Christian Huemer (University of 
>> Vienna) who with Birgit Hofreiter (and others) has published at least 
>> four papers containing the subject/topic in question. He is currently 
>> the Vice Chair of CEFACT's TMG.
>>
>> -Klaus
>