I agree with Michael on this. I think the question is, who decides and validates if the service changed. I think there should be an option to allow service consumers to be notified even when the service provider thinks there is only a semantic change because in most cases, it is the service consumer who will be held responsible.
H.Ozawa Michael Poulin wrote: > I have found two issues too, but in the James' answers. > > When dealing with a SOA Service, there are always 2 players - Service > and its Consumer. If a Service has changed, the Service side > (Provider) has to maintain service versioning. Period. The real > question is if the Service Consumer is fine with the change(s). i.e. > if the Service meets Consumer's requirements and the service change > results in forward compatibility for the Consumer. > > So, we have to have a mechanism which would allow a Consumer to be > sure that the version of offered SOA Service is the one the Consumer > needs. Let's assume a change in the algorithms used in the > implementation of the Service (not in the interface) which affects > accuracy of the returned results. The question is: whose job - Service > Provider or Service Consumer - is to validate that the change is > irrelevant for the Consumer A and B but it does not meet requirements > of the Consumer C any more? Answeri ng this question, please, consider > multiple Consumers with multiple SOA Contracts and dynamic nature of > the Service engagement... > > So, instead of thinking for everyone and run in almost endless > decision procedure (to inform the Consumer in particular case or not > to inform), I propose having a method and procedure for any Consumer > and relay on the Consumer's interest in the Service to check whether > the change requires special actions on the Consumer side. > > If anybody is interested in such mechanism, I have an article about it > (referenced several times already in this discussion). > > - Michael Poulin > > James Taylor <[EMAIL PROTECTED]> wrote: > > --- In service-orientated- > [EMAIL PROTECTED] > <mailto:architecture%40yahoogroups.com>, "jeffrschneider" > <[EMAIL PROTECTED]> > wrote: > > > > If the logic in a service is changed and f(x) begins producing a > new > > result, do you version the service? > > > > (Note: in this scenario, the interface didn't change just the > internal > > logic.) > > > > Thanks, > > Jeff > > > I think the question has two answers - if the changes impact the > business purpose of the service, yes, otherwise no. If the service > provides the same business decision it did before, no consumer should > need to worry. Check out this post for a little more on this: > http://www.ebizq.net/blogs/decision_management/2006/10/agile_services_ > <http://www.ebizq.net/blogs/decision_management/2006/10/agile_services_> > business_rules.php > > > ------------------------------------------------------------------------ > Do you Yahoo!? > Get on board. You're invited > <http://us.rd.yahoo.com/evt=40791/*http://advision.webevents.yahoo.com/mailbeta> > > to try the new Yahoo! Mail. > Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/service-orientated-architecture/ <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/service-orientated-architecture/join (Yahoo! ID required) <*> To change settings via email: mailto:[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
