RE: Highlights of the changes between BPSS 1.01 and 1.1

the 5 people that have carried all the work for BPSS 1.1 since January 2003 have resigned from the UN/CEFACT BPSS working group. All these people have elected to work to OASIS ebBP (note that what is important in the name of the TC is ebXML). My understanding is that the group will do whatever is in the best interest of ebXML to provide a business process/collaboration framework. I remain convinced today (after looking extensively at the web service technology) that ebXML is still the best model for a large class of B2B problems (maybe not the travel agent scenario or the loan approval one) but pretty much everything else. Now that does not mean that ebXML should not be layered on top of web services. It should ! And I think/hope this is what will happen.

The TC is not be bound anymore to follow something like UMM unless it is in the best interest of ebXML. On of the issue in the past four years is that we had to produce an exact subset of UMM, regardless of whether UMM fit in the ebXML architecture or not or regardless of its limitations if we needed to go beyond its capabilities. In addition, even though there seem to be a lot of good things in UMM, it has been a constant moving target and each time I had question, I had no one there to answer them. The UMM team does not use email to communicate apparently. Can't be more open for a process.

On August 21st, the UN/CEFACT and the TMG published a press release saying that all the ebXML work was complete. This would terminate the relationship with OASIS. Even if UN/CEFACT continues some work on the BPSS spec, I see two issues: 1) who will carry this work. Many more people have signed up to the OASIS ebBP 2) we can guess that no effort will be made by the TMG to align this effort with ebXML (CPP/A for instance).

You also have to consider that the process at OASIS is much clearer, democratic and open, I have never heard of such interference as there has been with the UN/DEFACT working group. Up until December 2002, our group had followed the UN/CEFACT process to the letter, at this point Klaus Naujok imposed a new project leader (Dave Welsh from Microsoft) and changed completely the work of the group, arguing that UMM 12.0 was just completed and we should immediately conform to it. Jim Clark from Microsoft had rewritten completely BPSS, changing pretty much everything (e.g. Binary collaborations where not there anymore !). Klaus and Dave Welsh at this point asked us to approve this proposal as is, nothing could be discussed. Later Jim removed his proposal (he was allowed to !) and we were allowed to continue. We just had lost 6 very important months ! I can't believe that anyone would qualify what has happened as an open process like the TMG is touting in its latest press release. I can't be sure, but I don't think this kind of things will happen at OASIS. I also don't think that anyone at OASIS will change the title, copyright notice and authors without prior informing the team as this has happened at UN/CEFACT TMG.

In summary, I think that with ebBP you have the guaranty that the work will be focused on ebXML and ebXML only, no hidden agenda, and that the process will be open for anyone to contribute. We actually welcome anyone to join the technical committee an be involved.