Kick-off European ebXML Interoperability Pilot Project

1
Kick-off European ebXML Interoperability Pilot
Project

• Brussels, 2002-09-25
• Pim van der Eijk

2
Overview

• CEN ISSS eBES Vendor Forum
• Background and status
• Report on Questionnaire
• Objectives, scope, capabilities, interests
• Moving forward
• Determine structure and timeline
• Make working arrangements

3
CEN ISSS eBES Vendor Forum

• CEN ISSS eBES workshop initiative
• Incompatibility of B2Bi systems
• Market needs for combined XML and EDI solutions
• Vendors need to give clarity on strategy to users
• SME involvement
• Lack of participation of European vendors in
  ebXML
• Defend European point of view in the global ebXML
  arena
• Promote participation in UN/CEFACT and OASIS work

4
Mission and process

• Vendor Forum Mission statement
• Mobilizing Vendor involvement in the
  standardization process, the awareness and the
  implementation of ebXML in Europe.
• CEN workshop process
• Similar to OASIS Technical Committee (TC)
• Delivers CEN Workshop Agreements (CWA)

5
Tasks and projects

• Tasks
• Education, awareness, standardization, technical
  assessment and migration
• Projects
• ebXML interoperability project

6
Status

• First meeting (March)
• Brainstorm session
• Initial idea for ebXML interoperability project
• Second meeting (June)
• Decision to focus on pilot
• Request to OASIS to endorse and support
• OASIS supports
• Internationalization of member base
• Facilitate access to technical work

7
Status (contd)

• Invitation letter to participate (August)
• States some participation requirements
• Some twenty respondents
• Questionnaire (September)
• Ten respondents
• Input for scoping discussion, reported on today

8
Today

• Kick-off meeting
• Decide on objectives and scope
• Understand each others interests and
  capabilities
• Form project teams
• Make working arrangements, including decision on
  commitment
• Requirements
• Vendors to commit to contribute resources
• Users to help provide business cases

9
Today (contd)

• Format
• Plenary presentation of results
• Interactivity, please !
• Make working arrangements
• Perhaps split up in groups for discussion
• Wrap up

10
Report on Questionnaire
11
Objectives of the questionnaire

• Understand the requirements of all participants
  before defining and starting the project
• Determine which ebXML modules vendors
• already implement in their products.
• want to get from other participants, or develop
  in this project.
• want to verify interoperability for.
• Get input on business scenario for demonstrator
• From (customers of) vendors
• From users and industry groups

12
Respondents

• CIDX / Solvay (BE)
• Cronos / XT-I (BE)
• Dan Net (DA)
• EAN (BE)
• Excelon (US/NL)

• Seeburger (GE)
• Software AG (GE/NL)
• Sun Microsystems (US/CH)
• TIE (NL)
• XML Global (CA)

13
Sections in the questionnaire

• Objectives and scope
• Level of support for ebXML modules in your
  products
• Priority for ebXML modules
• Other relevant software
• Case studies
• Payload (content) formats
• Liaisons
• Process and IPR

14
Objectives and goals score

1. Strongly disagree
2. Disagree
3. Agree conditionally
4. Agree
5. Strongly agree

15
Objectives and goals averaged
16
Difference from average
17
Disagreements

• Leading in Europe and Complementary
  products/services
• EAN ? has global (not European) focus, user
  organization
• CIDX ? not applicable, user organization
• Focus on mature parts, esp. ebMS
• Sun CPPA and BPSS as important as ebMS
• Cronos dont just work on ebMS, rest is
  important too
• XML Global start with the registry

18
(No Transcript)
19
Disagreements (contd)

• Limited resources, use (pre)sales budget (3.6)
• Seeburger (2) interoperability is first of all
  about implementation, sales/marketing comes next
• TIE (2) some participants will need to do a lot
  of technical work, implementations are far from
  plug-and-play yet
• Deliver demonstrator (4.3)
• XML Global (3) focus is to show that ebXML
  methodology is useful and relevant

20
Disagreements (contd)

• Migration scenarios (avg. 3.4)
• XML Global (2) work from registry, via CPP, to
  integration with back-end systems of real users
• Sun (2) focus on SME user not EDI user
• Find funding (4.1)
• TIE (2) first do the technical interoperability
  work, otherwise project will fail

21
Objectives and goals summary (tentative, my
interpretation)

• Start with limited scope and duration
• mainly marketing exercise for some
• but serious investment, technical effort for
  others
• Aim for a demonstrator that
• is a real-life case study
• is delivered and marketable in summer 2003
• demonstrates what ebXML is designed for
• might help a consortium apply for FP-6 funding
• Get a basic infrastructure operational quickly
• but understand that some of us are at an early
  stage still
• leveraging components from other participants

22
Support for ebXML scores

1. No implementation, or alpha version for internal
   use only
2. Partial implementation, not generally available,
   does not yet have the maturity of a product,
   usable under strict conditions
3. Implementation available, but not yet tested for
   interoperability, usable for initial pilot
   projects but not for production use
4. Reasonably complete, usable in end-user projects,
   but interoperability not yet verified
5. Complete, commercially available implementation
   that has passed interoperability / conformance
   testing and is being used in production settings.

23
Module support average
24
Module support difference from average
?
25
Module support summary

• Participants in all three categories
• Advanced, intermediate, just starting
• Emphasis clearly on messaging and collaboration
  protocols
• Other modules supported by specialists
• Participants collectively implement significant
  parts of ebXML already !!

26
Module priority scores

• Unimportant, the project should not waste its
  time looking into this
• Not important, but nice to have
• Moderately important, if possible the project
  should have a look at this
• Very important, the project should make an strong
  effort to be able to provide this
• Essential, the project is not a success unless
  this module is implemented

27
Module priority results average
28
Module priority results per vendor
29
Module priority summary

• Broad agreement that focus should be on ebXML
  messaging
• Configured by Collaboration Protocol Agreements
• Much less agreement on registry, core components,
  BPSS
• Vendors that are further ahead tend to be more
  cautious CC and BPSS

30
Other software

• CIDX CIDX tools
• Cronos transformation
• Dan Net transformation
• Excelon BPM, modelling
• Seeburger EAI, transformation, workflow
• Sun UBL concepts
• Software AG XML server, EAI, transformation
• TIE integration, transformation, forms
• XML Global Security, transformation, message
  store, payload generation

31
Case studies
32

• We wish that this project is very closely
  aligned to real world issues, and so test
  scenarios and payloads and such should come
  directly from end customers and vertical
  organizations (Sun)
• It is important that this is a real world
  demonstrator and as such the ebXML aspects must
  be interfaced to demonstrate a real end-end,
  possibly multi-party, system (TIE)

33
User participation today

• CIDX
• Chemical Industry XML B2B standard
• Raymond Betz (Solvay, Brussels) has offered to
  discuss a potential case study
• EAN
• Involved in OASIS interoperability work
• EAN UK (e-Centre) has an ebXML project under way
• Bolivar Pereira (EAN International, Brussels)

34
Other suggestions

• From participants
• Seeburger Papinet
• Software AG Accord, Opentrans
• Sun have a high-level scenario Sabre?
• At a later stage
• CEN EEG9 (healthcare informatics)
• Government and tax agencies, XBRL
• Some other large multinational companies

35
Message payload scores

1. Irrelevant
2. Somewhat important
3. Important
4. Very important
5. Essential

36
Payload preference average
37
Preference variation per vendor
38
Payload format summary

• Clear preference for UN/EDIFACT
• EAN-UCC XML as next best
• No strong preference for other formats
• Plus, lack of agreement
• Decision on payload format is dependent on
  business process and should be based on the
  demonstrator scenario

39
Liaisons
40

• OASIS IIC TC interoperability guidelines
• Drummond Group for UC-Council (commercial)
  reference implementation
• Second series completes in November
• Other ebXML pilots worldwide
• Relevant project teams in UN/CEFACT and OASIS

41
OASIS ebXML Implementation, Interoperability
Conformance TC

• Chaired by Jacques Durand (Fujitsu)
• Work on interoperability guidelines and automated
  test facilities
• Now moving to industry deployment (templates for
  EAN-XML, RosettaNet)
• Want to coordinate various initiatives
• US (Drummond Group, OAG/NIST), Japan (ECOM),
  Europe

42
OASIS IIC (contd)

• Main input to our project
• Basic ebXML interoperability guidelines
• Working on
• core test suite synthesized from various
  initiatives
• Definition of interoperability profiles
• Template mechanism for project-specific
  extensions
• Generalized from EAN input
• Deliverable from our project to IIC
• A deployment template for EDIFACT?
• Liaison via Steve Yung (Sun), Bolivar Pereira
  (EAN)

43
Other liaisons

• OASIS ebXML Messaging TC
• Ian Jones, chair (BT, UK)
• Expressed willingness and interest to support
• ebXML joint coordination and marketing teams
• Research projects
• Started talking to OpenXchange
• Other DG Information Society projects

44
Participation in interoperability projects
45
Process issues
46
Process

• Vendor Forum is a project of eBES
• eBES is a CEN ISSS workshop
• By default, should adopt CEN process and IPR
  policy
• OASIS endorsed and supported
• Joint membership condition
• CEN, OASIS to work out formal relation

47
Intellectual property

• According to questionnaire, there are no
  objections to royalty-free license
• This is ebXML
• Vendors will retain all rights over their
  products
• No automatic endorsement by others
• Suggestion transfer IP to OASIS IIC
• So there is one place maintaining this
• Discussion topic for CEN ISSS and OASIS

48
Moving forward

• Here the discussion should become very
  interactive

49
Proposal Two main work areas

• 1 Messaging interoperability related
• Bottom up, technical focus
• Start with ebMS and ebCPPA
• Sort out infrastructure issues
• Leverage related projects and experience
• 2 Business-oriented, demonstrator related
• Top down, business process-oriented
• Start with UML use case, activity etc. diagrams
  towards BPSS collaborations, business
  transactions
• Adopt an existing payload, e.g. EDIFACT

50
XML Global registry-centric approach

• Start with ebXML registry hosted by CEN
• Participants start adding CPPs of real
  organizations
• Participants implement BSI for those CPPs
• Some CPA negotiation ?
• ?
• Offering XML Global Registry free of charge
• ebRR 2.0 compliant implementation
• Can be viewed as 3rd work area, orthogonal to
  other two?

51
Area 1 Interoperability testing

• Leverage related projects
• IIC, e-Centre, Drummond Group
• Participants involved in both (EAN), Excelon,
  Sun, XML Global
• Select or define test suite
• IIC testing framework and suites
• Internal test suites of participants?
• Messaging profile
• Infrastructure
• Computers, network, software, firewalls, ebXML
  MSH send and receive, ping and pong
• CPPA configurable

52
Interoperability testing (contd)

• Here we need input from participants with
  experience
• Practical issues
• Do we need central infrastructure? (hardware, OS,
  network etc.) or can this be done remotely?
• What are the steps to take?
• Some simulated back-end, replaceable by real
  integration tools

53
Interoperability testing (contd)

• Need for two speed project
• Participants with advanced, conformant
  implementations
• Participants that have initial / prototype
  implementations only
• Need a way to support both

54
Area 2 Demonstrator definition

• Decide on final project deliverable
• Work back from there and determine steps to take
• Which business case?
• Process, information, parties involved
• Payload format
• Provided by which participants?

55
Project teams

• This would give (at least) three initial teams
• ebMS team, advanced
• ebMS team, catching up
• Business case team
• Teams should have their own meetings / conference
  calls in addition to the plenary ones

56
Break out in groups

• ..

57
Tasks

• Write charter, project plan
• Set up conference calls, mailing lists etc.
• .