← Prev in month
← Prev in thread
Next in thread →
Next in month →
RE: [ebxml-msg] roles o f IIC and of Drummond Group
Title: RE: [ebxml-msg] Have you implemented Version 2.0 ? Lisa and all: Let me report on the general view and line of action of the ebXML IIC TC on this - as this was questioned, (and we have had this debate quite some time before you did :) First, in the IIC TC we are currently designing - with intent to develop as well - a conformance testbed, for testing the conformance of an MSH implementation to the MS standard. I am chairing a small task force to this effect (including NIST, XMLGlobal, Fujitsu). We certainly welcome any additional contribution (so far my calls have been limited to the IIC maillist) Now on the interoperability / conformance issue: When we started to look into this at IIC (last August), we came to realize that: (1) Conformance to the spec does not guarantee Interoperability. Indeed, interoperability is more about compliance to an "agreement" (formal or not) than to the specification, and is always dependent on many factors that are out of the scope of a specification : reliance on and features of 3rd party packages (e.g. security), transport issues and configuration, level of XML validation, and many other mundane details that you cannot imagine unless you have actually been involved in one of these past ebXML POC demos. Currently, we recognize that the only sure way to guarantee interoperability between diverse implementations, is to do what DGI is doing. And we have to recognize that with the tools of today this is quite resource intensive (though we certainly hope - and act- so that it will not stay that way!) and is currently way beyond the resources of IIC. The way the IIC will contribute to interoperability, is primarily by releasing MS Implementation Guidelines (currently being written). This alone will clarify and remove many ambiguities, inform on tools - or suggest implementation choices, take a stand on "gray" spec areas, which in turn will reduce possible mismatches. (2) Interoperability is not a substitute for Conformance. Just because you have a group of implementations that can interoperate 1-to-1 does not mean they are conforming to the specification. They might indeed restrict themselves for convenience to a subset of features, or a to particular modes of operation, or share the same "interpretation" which can be too narrow or flawed. This is more in the IIC charter to check this, we believe. So we are designing conformance test cases, and a testbed. Our first objective will be to implement end-to-end a small portion of conformance testing, and to make this as automated as possible. (building an exhaustive MS conformance test requires also a substantial effort - we expect to attract more resources after a first demo!) It is too early to tell at this point whether and when we will have a *full* conformance test procedure, something a certification authority can use. But we are careful to develop a testing technology that: (1) can be extended to various and additional test cases, (2) is reusable for future versions of MS (and even different messaging environments) (3) can be used to support interoperability testing (4) is automated (XML-based test driver and test cases, separate validation of the trace, etc.) The IIC has no working or business relationship with DGI (except we share a member...), but we acknowledge their role. However we (IIC) participate in the OAG B2B TestBed activities, helping OAG (www.openapplications.org) and NIST to upgrade their current HTTP testbed to ebXML. There is synergy here with IIC - both sides can benefit from each other experience. The general scenario we envision that provides the rationale for all this is: 1- Company A has implemented an MSH. It will first undergo conformance testing. We expect this can be done with little external assistance, by remote access or even downloading the testbed. (However, if certification is the objective, a certification authority, e.g. NIST, will need to validate the test outputs). 2- Company A will then undergo interoperability testing with other MSH implementations (e.g. using DGI tests). This should be easier/faster (and therefore less expensive), given that the previous conformance test has removed many roadblocks. 3- COmpany A can then deliver its MSH to customers along with its B2B products, or use it itself for business. The user company will then participate to application-level B2B testing activity as required by content standards such as OAG or RosettaNet, focusing on the business content of the message, and the choreography. 4- Company A and its peers still perform occasional "maintenance" testing procedures, using test cases and technology developed in (1) and (2), as this is critical to their business. So everyone is putting resources in this, and certainly the investing companies all have some form of interest in the outcome - possibly not as "fast-return" as in the case of DGI-, but nevertheless worth it in the longer run. We also believe the testing technology and know-how that goes with it - interop or conformance - is critical to the success of such business standards, and is something that goes beyond one-time validation, as these environemts are doomed to evolve and undergo separately multiple configuration changes and upgrades. And to conclude on a practical note, remember we welcome contributions on MS conformance test cases and testbed... Regards, jacques Durand ebXML IIC Fujitsu -----Original Message----- From: Ralph Berwanger [mailto:] Sent: Friday, March 01, 2002 12:53 PM To: Cc: Bill Morgan Subject: RE: [ebxml-msg] Have you implemented Version 2.0 ? Lisa, Before we start beating on Drummond Group we should be careful to understand what they were offering. They were not certifying that vendor products were 'compliant' with the specification. They were offering a service to ensure interoperability between products of various vendors. This may seem like hair-splitting (and for those who know me you understand how difficult that is), but it is the simple truth. Vendors were not forced to engage in the testing, they voluntarily participated. There was NO other venue and at least this method provided a somewhat impartial third-party validation of facts assumed by the vendors. I agree that it would be improper to take intellectual property 'donated' by various organizations and sell it for gain. Like you and the rest of those on the listserv, we spend significant resources assisting with standards work. It is intended to produce an open consensus-based standard, not to advance the interest of any individual company. I think we are sharing the same soap box. However, in fairness to the Drummond Group, that is not what has happened. They took the specification and produced a test plan which allowed the participants in the test to exchange messages and to systematically exercise their solutions to ensure that they worked with similiar products from other vendors. In the end, those that participated can affirm that their products can interoperate. They might even say that their products are compliant, but not based on the results of the test. That was not the purpose of the test. For the record, I do not totally agree with the statements that David has made below, but I cannot fault Drummond Group for their actions. Regarding compliance testing, a very different process is needed. I exchanged messages with Prakash yesterday on this topic. I am hoping that the OASIS IIC TC can move this forward. I am prepared to assist anyway necessary to help this work progress. I think we may need to bring additional resources to the table to assist the IIC. Additionally, there other efforts, like the B2B Testbed, being proposed by OAG and NIST that can provide additional support to this area. Ralph Berwanger Ambassador to Standards bTrade,Inc --------------------------------------- -----Original Message----- From: David Fischer [mailto:] Sent: Friday, March 01, 2002 2:05 PM To: Seaburg, Lisa; Subject: RE: [ebxml-msg] Have you implemented Version 2.0 ? DGI has donated thousands of hours to ebXML over the last two years. We are represented on five committees (two as editors and one as v1.0 Chair) all non-reimbursed. The interoperability test ran about four months with daily conference calls and constant work on our part. We are providing this service for very little more than cost recovery. DGI is not taking advantage. Why is this any different than someone implementing the spec and selling their product, thus taking advantage of hundreds of hours of volunteer work? No one actually volunteers for ebXML. Everyone has a business reason to be here. Regards, David Fischer Drummond Group. -----Original Message----- From: Seaburg, Lisa [mailto:] Sent: Friday, March 01, 2002 1:04 PM To: Subject: RE: [ebxml-msg] Have you implemented Version 2.0 ? Excuse me, but if this is true, the fee is $35,000! Then the Drummond group is blatantly taking advantage of the hundreds of volunteer hours put in by all the people involved. I would like to get on my soapbox about this, but I won't today. I will wait and see what other comments come from this email thread. Lisa -----Original Message----- From: Duane Nickull [mailto:] Sent: Friday, March 01, 2002 7:58 AM To: Patrick Yee Cc: Cliff Collins; Subject: Re: [ebxml-msg] Have you implemented Version 2.0 ? The stated charge is a bit steep at $35,000 a pop. Duane Patrick Yee wrote: > > Excuse me, I want to ask a side track question. > What is the charge to have the messaging implementation to be tested by the > Drummond Group? Thanks. > > Regards, > -Patrick > -- > Patrick Yee > System Architect > Center for E-Commerce Infrastructure Development (CECID) > Dept. of Computer Science and Information Systems > The University of Hong Kong > Tel: (852) 22415674 > Fax: (852) 25474611 > > ----- Original Message ----- > From: "Cliff Collins" <> > To: <>; <> > Sent: Friday, March 01, 2002 2:06 AM > Subject: RE: [ebxml-msg] Have you implemented Version 2.0 ? > > > Sybase has implemented Version 2.0 of the specification. Our > > product is also part of the interoperability test that > > is being run by the Drummond Group. > > > > Cliff > > > > > -----Original Message----- > > > From: [mailto:] > > > Sent: Thursday, February 28, 2002 9:47 AM > > > To: > > > Subject: [ebxml-msg] Have you implemented Version 2.0 ? > > > > > > > > > Implementers > > > > > > As I only have a couple of outstanding non returned ballots the > > > specification looks likely to pass I need confirmation that the > > > specification has been implemented and does work as we have specified > for > > > the submission process. I would like this to be open so please post an > > > e-mail to the list confirming you have successfully implemented > > > Version 2.0 > > > of the message services. This is NOT a product announcement only that > you > > > have implemented and tested it internally. If you do not want this > public > > > please e-mail me direct and only the number of implantation will be made > > > public. > > > > > > Ian Jones > > > Chair OASIS ebXML Messaging Services TC > > > > > > Tel: +44 (0)29 2072 4063 > > > Fax: +44 (0)29 2072 4137 > > > Email: > > > > > > > > > ---------------------------------------------------------------- > > > To subscribe or unsubscribe from this elist use the subscription > > > manager: <http://lists.oasis-open.org/ob/adm.pl> > > > > > > > > > ---------------------------------------------------------------- > > To subscribe or unsubscribe from this elist use the subscription > > manager: <http://lists.oasis-open.org/ob/adm.pl> > > ---------------------------------------------------------------- > To subscribe or unsubscribe from this elist use the subscription > manager: <http://lists.oasis-open.org/ob/adm.pl> -- CTO, XML Global Technologies **************************** Transformation - http://www.xmlglobal.com/prod/foundation/ ebXML Central - http://www.xmlglobal.com/prod/central/ ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl>
← Prev in month
← Prev in thread
Next in thread →
Next in month →