OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

ebxml-core message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: RE: Syntax Free Models


I agree in terms of process. I am assuming that when we model processes and
choreographies, that we allow for real-world variation in our models.
Describing this range of variation in a controlled way is easy to do with an
XML syntax, and seems to be very much the approach taken in XMI and similar
efforts. I did *not* mean to suggest that we can hard-code absolute
sequences into our business processes.

Some workflow systems - I speak from experience with the one sold by
Documentum - manifest the process as an object that can be modified at any
point during the process it describes. This is a handy way of accomodating
branching and variable sequences within processes, and is very common in
workflow technology. What we do should allow this kind of technology to be
applied between trading partners, rather than simply within the enterprise.

My assumption is that, in terms of an XML implementation, the choreography
would be manifested as a document that indicates both the steps of that
portion of the business transaction that will occur between trading partners
(rather than the portions internal to the enterprises) for the particular
transaction; and an indication of where in the process a particular document
is placed. These functions - indicating the choreography as opposed to
indicating the state of a given transaction - could also be done separately,
if that makes sense.


Arofan Gregory 

-----Original Message-----
From: Bob Haugen [mailto:linkage@interaccess.com]
Sent: Wednesday, March 01, 2000 9:19 AM
To: ebxml-core@lists.oasis-open.org
Subject: RE: Syntax Free Models

Arofan Gregory wrote:
>Process descriptions - because they model movement in time - are
>tightly bound up with sequence. Data descriptions, because they are not
>bound up by time (other than indirectly, by referenceing a sequence
>description as part of identifying their context), should be left

I agree with you about data descriptions.  

Process descriptions, while they must deal with sequence, 
cannot have sequence relationships "hard-coded".  Manufacturing
and distribution processes, for example, may vary according
to the presence or absence of available inventory, the properties
of process outputs (off-spec requiring rework, potency of active
ingredients, sweetness and other recipe properties) and many
other process constraints that act on the "event horizon".
(I expect you meant that, but want to make sure.)

Another situation where sequence can be important (that Martin
focuses attention on) is the conversation or choreography or
interaction of messages between parties in for example 
making agreements.  There, too, the sequences cannot be
hard-coded, but are important.

Bob Haugen

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help

Powered by eList eXpress LLC