On 1 October, David Sprott of CBDI has published his observation of 'Navigating the SOA Standards Landscape Around Architecture', under the title "The Failure of SOA Standards!" You have, probably read my report in this topic back in July. The leitmotif of David's post is "Currently we see minimal use of these concept standards because they are inadequate. We recommend that you continue ... using the CBDI-SAE meta model. " It's 'kinda' familiar song, isn't it?
The major argument provided is that the SOA standards that all three The Open Group, OMG and OASIS work on are too abstract to drive SOA code generation. Well, may I kindly ask what code is meant for Architecture?
As David has admitted he had "more respect for the OMG's work in SoaML, because it is usable in support of code generation". Is it because CBDI has "actively participated and contributed to the OMG SoaML work and we [CBDI] are pleased to announce a high level of alignment. This is possible because of the highly generic nature of the SoaML work"?
I can confirm that OASIS SOA has also contributed a lot into SoaML, which is resulted in a lot of right words in correct order have appeared in the standard. But, as I mentioned in my another post, it seemed it was not enough to make that standard edition really service oriented. To me, current version of this standard better be named "sOaML" because its orientation has been realised as role/participant-Oriented rather than orineted on the service architecture.
It is a pity that so talented team that moves toward Business Service Model (which is lovely to my heart) has compromised their efforts by requiring a code generation for architecture and judged other standards from this perspective. For instance, while I respect the CBDI concept of Business Service Model, I do not necessary need it because the OASIS SOA RA (coming public review) is enough for me to extend SO architecture into Business.
Leave a comment