Minutes of 10-05-05 session in Hangzhou

[Mavis Cournane <mavis.cournane@cognitran.com>]

Dear all
please find an account of today's session in Hangzhou.

Regards
Mavis

----------------

Minutes, Hangzhou UBL F2F 10-05-05

Attendees: Tim McGrath, Mavis Cournane, Yukinori Saito, Mikkel Brun, 
Anne Hendry, SungHyuk Kim, Thomas Lee, Peter Borresen, Colin Lam, 
Wenfeng Sun

1. ESLSC Report (Tim McGrath on behalf of Oriel Bausa  of ESLSC)

ESLSC covers both Spain and South America.
They have gone through a UBL 1.0 translation. They have built software 
libraries for UBL in Spain. There are packages in both COM and Java.

One of their recommendations is to deal with Invoices. This is similar 
to the Danish recommendation.

The UBL Code Lists did not contain everything that was required.

200 Companies are using the UBL libraries to generate the invoices. 
Between now and June they will be going live with 10 companies sending 
real invoices.

In South America there is an Ecuadorean working group. There is a 10 
month plan to define the UBL data needs, the pilot project and the 
dissemination plan.


2. Review of the International Data Dictionary (IDD)
We now have a document that is a UBL CD that will require maintenance 
and we will need to manage the maintenance of this. We need a 
synchronization plan.
We should not break backwards compatibility but there will be a need to 
translate new entries and to improve translations of existing entries.

We need to ensure coordination between the various localization SCs.
Saito-San said that the meeting in Copenhagen was the beginning for 
this coordination.

In Japan individual translation was done, and then a peer review took 
place.

In Korea this was done more in a group with a coordinator.
In China, there were 3 phases. First the BIEs were translated, one 
person did the draft translation, it was group reviewed, this was just 
within CNIS this review.
During this process much attention  was paid to the old EDI 
translations. The existing translations of EDI were then reviewed by 
Industry Experts.
In Japan existing EDI translations were also used as basis for the 
initial translations.

WS: Regarding the Chinese translations one would expect translation 
problems with the BIEs that were not part of EDI.

MC: We should try to steer this issue
a. Recommendations for new Localization SC
b. Changes to the process required for maintenance
c. Changes to the process for translations to the new dictionary entries

Summary

Generally, an expert would be appointed to perform an initial 
translation. Attention should be paid to any existing translations of 
BIEs (e.g. EDI translations). The initial translation should then be 
reviewed by a group of experts who will look at harmonization and 
semantic integrity.

MC: As we move forward we will have difficulties maintaining 
consistency/integrity in both the source and target dictionary entries.
TM: This is why an Ontology is needed.
The problem we have is that we have a dictionary that is a living 
document. There will be extensions and revisions to this. Any one of 
the components of a dictionary entry can change at any time which might 
trigger a need to review or change in the target.

MB: Another way of validating the translations is to see that the data 
when translated is formatted in the right way. For this you would need 
a stylesheet in the different languages. These formatted views could 
then be used by the LSC review teams to review the translations.

TL: We will also need to translate additional metadata, for example, a 
label for display.

TM: We need to create a project team for this.
The way to progress this would be to ask HISC for help.
We need to have an abbreviated label for the output media.

One of the attractions of the IDD is that it could be a repository on a 
website and the entries would be URIs that could be pointed to.

Action:

Tim will take responsibility for proposing a plan to get the team 
established.

The environment for translation will be set up, but we won't apply 
translation until CD 2.0 is finalized.

AH: There will be a requirement to do some versioning as there will be 
cycling between translation teams and the IDD creation team where 
clarifications will be sought to the source that impact translation. 
Individual items should be versioned.

TL: There is a difference between the versioning and telling what has 
changed.
TM: We can use the versioning as an indicator for the change log and do 
two things at once. Every entry has to be versioned so it is readily 
identifiable what has changed.
Similarly, the target entries will be versioned.

TM: We are not aware of having received any external comments on the 
IDD (except for compliments).

3. New UBL issues
These have been added to the UBL issues list. Mikkel Brun and Peter 
Borresen (Danish Interest Group) will add the IDA/OGC issues to the 
list tomorrow.

4. Extended Process Model

Sourcing process diagram

TM: One of the challenges is to think up a better document name than 
catalogue.
Are we in agreement on the use case for the sourcing activity diagram.
MB: I think the use case is misleading for how the catalogue is used. 
The most common use case is for a seller to send off a catalogue to 
various market places. He could send it to a buyer but that would be 
rarer.
Saito-san: In Japan the catalogue is not often necessary.
TM: THe catalogue part of the process should not be emphasised, and 
should not be an integral part of the process. The request for 
quotation and the quotation are an independent and important part of 
the process. The catalogue question is important between the seller and 
the marketplace but not between the seller and the buyer.

MB: The Danish government would be interested in sponsoring the 
development of cataloguing document that extends the current proposed 
process.

Agreed:

The Source process diagram needs to be re-drawn to add an additional 
swim lane for seller providing a catalogue to the  marketplace 
(broker??) and the marketplace (broker??) provides the catalogue to the 
buyer.
We should split it out in to two diagrams one to show the sourcing and 
the other would be quotation.

Action: Mikkel Brun will do this.



Order Activity Diagram

No change

Fulfillment activity diagram

There is a proposal to have a rectification. The current process works 
on the idea the seller can send someone a despatch advice and you get 
back a message saying this is what I have received. There is a 
possibility that these two do not match. How is this discrepancy 
resolved i.e shortfall or overfill. THe proposal is to have a 
rectification advice to help reconcile the difference. This 
synchronization allows the billing process to begin.

Saito-san: Rectification advice is not a word that is a word that is 
recognizable. A better word would be "Correction Advice"

PB: Would this need to be a new document. You just resend a corrected 
despatch advice.

Action: PB will explore the idea of having just a new despatch advice. 
Should there be just a New Receipt advice as well. Can a new despatch 
advice satisfy the requirements of a rectification advice?

Buyer Billing activity diagram

We need to know more about the information flow before a self bill 
invoice is created, for example, how does the buyer understand what to 
provide in the self bill invoice.
Action: Peter Borresen will mail the IDA/OGC to find this out.

Seller Billing activity diagram
PB: There is a document missing from the diagram - Account Response.
It is also missing a loop.
TM: This needs to be redrawn

Action: Tim to redraw this diagram


Payment activity diagram
There is a requirement for an acknowledgment at the business level.
Where an invoice is rejected a message is required to say that the 
invoice has been rejected (Invoice Rejection message).

Action: MB will redraw this diagram to include the invoice rejection.