Asuman,
Both this example and the one Stephen posted exhibit similar needs. Format A, Format B, similar domain characteristics - need crosswalk between them.
Conceptually I'm seeing the following approach can work.
I will try and work something up on this over the next few of weeks - but my plate is very full at the moment.
Conceptual approach -
Step 1 - map targets into abstraction format
Step 2 - align to common terms dictionary (this can be part automated and part manual)
Step 3 - consolidate into use templates for each target
Step 4 - parse templates creating mapping A->B and B->A based on semantic information in templates
Notice that quite a lot of this is already available in the current CAM toolkit.
Mechanically this translates to:
Format A XSD -> ingest to CAM template
Format B XSD -> ingest to CAM template
Run XSLT against CAM CXF abstraction XML and compare to master dictionary - matching on names - generate crosswalk (we have this available for NIEM.gov example - would need to create similar dictionary for UBL and other domains).
XSLT -> CAM -> dictionary -> crosswalk XML (that's also can be opened automagically as Excel spreadsheet)
Have an example I can share using OASIS EDXL and NIEM dictionary and Excel
(I'll post something on that - so you can all try it yourselves - assuming you can drive XSLT).
Integrate results of matching crosswalk back into CAM template in <Extensions> section - with match to master dictionary called out for each element and attribute.
Crosswalk + CAM -> new template (manual editing)
Adapt the existing XSLT code that parses XML together with a CAM template - to instead generate simple XSLT mapping code (actually this is something Martin's been talking about is generating XSLT validation code... but that's another whole topic - we actually have an XSLT validating version of CAM in beta too - lots of xslt code here we can re-purpose to SET tasks!).
I can mock up a skeleton on this - proof of concept - then see what else is needed at that point. Where it gets interesting is how smart you can make the matching and reformatting... I'll start with dumb, with manual intervention for those things that don't match easily.
As usual you can get 80% of this working with 20% of the effort. I'm signing up here to do the 20% first! ; -)
At least after this exercise we can all see where the challenges are!
-------- Original Message --------
Subject: [set] Groups - An Example Use Case
(OASIS.SET.TC.METU.UseCase.doc) uploaded
From: asuman@srdc.metu.edu.tr
Date: Fri, December 05, 2008 8:03 am
To: set@lists.oasis-open.org
The document named An Example Use Case (OASIS.SET.TC.METU.UseCase.doc) has
been submitted by Ms. Asuman Dogac* to the OASIS Semantic Support for
Electronic Business Document Interoperability (SET) TC document
repository.
Document Description:
iSURF Interoperability Service Utility: Providing the Interoperability of
CPFR Electronic Business Documents
View Document Details:
http://www.oasis-open.org/committees/document.php?document_id=30296Download Document:
http://www.oasis-open.org/committees/download.php/30296/OASIS.SET.TC.METU.UseCase.docPLEASE NOTE: If the above links do not work for you, your email application
may be breaking the link into two pieces. You may be able to copy and paste
the entire link address into the address field of your web browser.
-OASIS Open Administration