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

 


Help: OASIS Mailing Lists Help | MarkMail Help

ubl-ndrsc message

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


Subject: NDR SC Minutes May 21 2003


Dear all
please find attached a copy of today's minutes. Please let us know next week
if you cannot attend the call.
Guther has put forward 3 discussion points under item 6 on the agenda which
we wish to go through next week

Regards
Mavis
--------------------------------------------

Roll Call and Welcome from the chair (Mavis)

Attendees:

Mavis Cournane
Lisa Seaburg
Tony Coates (left X:55)
Paul Thorpe
Kris Ketels
Ann Hendry
Gunther Stuhec


Regrets
Mark Crawford
Eduardo Gutentag
Mike Grimley
Eve Maler
Matt Gertner.

Others
AWOL

2.  Acceptance of minutes from previous meetings, we still need London
minutes finished.
quorum was not reached.

3.  Adoption of agenda/schedule planning

Upcoming Call Schedule:

May 21 - Mavis, chair (Decisions voted)
May 28 - Mark, chair (Editorial Work)
June 4 - Lisa, chair (Editorial Review)
June 11 - Mark, chair (Editorial Review)
June 18 - Mavis, chair (NDR 1.0 & Code List SC approval)
June 25 - Lisa, chair (NDR 1.0 and UBL CL Comm. Spec)
July 2 - Mark, chair (Clist & NDR 1.0 OASIS review)
July 9 - Mavis, chair (Clist & NDR 1.0 OASIS review)
July 16 - Lisa, chair (Clist & NDR 1.0 OASIS review)
July 23 - Mark, chair  (Clist & NDR 1.0 OASIS review)
July 28 - Aug 1 - Face to Face Montreal (Process OASIS NDR/Clist review)

4. Review NDR release schedule/action items:

20030514:  Mavis get last two days of London minutes sent out.
Outstanding, awaiting feedback.

20030514:  Lisa add double check NDR rules for business document naming as
agenda item for QA of NDR document.

This has been done and discussed. It is on hold because business document
naming is not something for the NDR. CSC will deal with file naming. The
actual names of the business documents will be handled by LCSC.

Ann Hendry, I thought they needed some guidance from us.;

20030514:  Start issue thread on email.  Ann Hendry, is grouping the
comments from the comment review and will send out each one as a separate
issue.

AH I have 8 of them that I have pulled out.
MHC: Let's go through them now.

1. How will UBL be reverse engineered to CCTS. There are 3 parts to it. See
comment 3.1 in the comment spreadsheet.

2. Document naming convention policy - will it be linked to namespace. They
mean the business document.

LS There is already naming rules through the dictionary entry. The comment
was the names in the spreadsheet were not consistent. It is the names within
the spreadsheet e.g OrderResponseSimple and elsewhere names have spaces in
them.
LS: THis is an LCSC issue and should be using the naming rules correctly.
The rule is defined in the ebXML CCTS 1.9
GS: LCSC has to look to the CCTS for these naming conventions.
AH: What is the NDR rules for then.
GS: THe LCSC should not look to the NDR rules because they are for the XML
schemas only.

3. See 3.5. What does the prefix CCD mean in the schema when ID="CCD". It is
a temporary identifier of the specific core component. It should be CCT.
Gunther Stuhec will check this out and get back to Ann Hendry on it.
GS: All temporary identifiers are not agreed upon yet. We should delete
these for now.
AH: Should that be done for 8.0
GS: Yes it is.
MHC: We need to think of some rules for it.
GS: Yes but after 8.0.
GS: The rules must come from LCSC as they are responsible fro semantics. NDR
is responsible only for rules of the generation of the XML schemas. Such a
rule would be for library content only.
The LCSC should define business document structure according to CCTS. We
take this library and we generate the schemas for it using NDR rules.

4. See 18.1: Digital Signatures.
We have to use binary objects for digital signature.  The binary object
allows the use of base64. Using the CCT BinaryObjectType you do the digital
signature.
PT: ASN1 is often used for these. The key to digital signatures is that
distinguished encoding rules. For any given piece of data there is one way
to encode it, to check the signature you re-encode it and check that the
signature matches.
AH: How does this differ from DER.
PT: DER is mostly used for signatures and is canonical all the time.
LS: This needs to be put forward to the NDR as proposal.
PT: I will put something together proposing of how this could be done.

LS: I have spoken to Dave Burdett about this. Where is the digital signature
placed in the document.
PT: The part of the document you want to sign you encode separately into
DER.
LS: We are only building XML not binary stuff.
We should make it generic enough so that people can handle it whatever way
they want.
AH: The intent of the comment is to allow digital signatures. LCSC did
decide this was a good idea.
The signature element should be optional at the end of each document rather
than at the envelope level.
People decided it should be ok to put it in the body. NDR needs to provide a
schema rule for this.

PT: Digital signature requires, the entire contents of what you sign must be
preserved as is for verification purposes. Since XML is not canonical in
itself you have to preserve the contents of the document with some other
encoding rules. For example Distinguished Encoding Rules.
TC: The Digital Signature Group took over canonical XML so that they would
sign the canonical version of the document.
PT: Some of our group in ASN1 identified some problems with this.
MHC: PT will send something to the list and Eve will be involved in the
discussion when she comes back.

5. See: 23.1 XML schema rule regarding parnter IDs and security.
This will be dealt with under 18.1

6. See B2: Common Aggregate Namespace is bloated.
AH: The intention is to have one virtual library of reusable components.
At the F2F it was asked what was the technical constraint on putting
everything in one reusable library.
GS: There is no constraint.  We need to decide how many namespaces we want
and definitions for them. Bill's paper does not go any deeper.
MHC: What is the priority for it.
GS: It is a key question.
MHC: I will talk to Bill and see if he can give us any input on it.
AH: This is not an issue for 80 but for 1.0

There are 3 more to come. We will do them next week.


20030514:  UBL chairs will have paper regarding the naming convention to be
used, for naming business documents.
In progress

20030514:  LCSC should be responsible for the Abbreviations and Acroynyms
Control List.  Yes, we agree.  Mavis and Lisa will send email of discussions
to the LCSC list.

TBD
20030514:  Mavis to go through minutes, look for any decisions made and have
list of rules that need to be reflected in the minutes.


IN progress and to be done by end of weekend.

20030514:  Reverse engineering of the schema. The NDR QA team should be
checking to see of the Perl script has any rules in it that are not
documented in the NDR document.  Also check for current NDR rules to make
sure they are reflected in the schema.

We are not checking the script as much as the schema. The script is being
investigated to see why some things we don't understand appear.

200305014:  Lisa will send Code List document to Tony Coates, and they will
continue conversation on code lists.

5. Discussion Items:

 Digital Signatures

Paul Thorpe will write something up and we will get input from Eve Maler on
this.

 Code List Rules document - Update


6.  Other Business:
GS: What happens if we can't put the qualifiers into the XML tag names.
IT is not possible to differentiate and build different structures in
different schemas in the aggregations.
MHC: THis is a question that should be added to the agenda next week.

GS: WHat happens if we can't define codes locally.
MHC: But in London we decided that this would be possible.
LS: We decided this in London and it will be reflected in a new version of
the code list paper.

GS: In London we decided that ObjectClassTerms cannot be used in the tag
names. THat means that PartyAddress cannot be used it has to be Address. But
you may have two different structures dependent on the ObjectClassTerm, so
we cannot differentiate the aggregations in the schema.
We can't modify the perl script to take account of the new rules until these
issues are dealt with.

Next NDR QA Team teleconference call:  NDR QA: Jack, Mike, Lisa, Kurt, Matt,
Kris
23 May 2003 - Friday: 8am PST:

7.  Adjourn y:15





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