Ted Stockwell wrote: > ----- Original Message ---- >> From: Niclas Hedhman <[EMAIL PROTECTED]> >> Subject: Re: [qi4j-dev] DDD sample implementation idea. >> >> Another question related to this particular case; >> Should we do the "RIGHT" thing or the "EQUIVALENT" thing?? >> >> If we go "RIGHT" we have a superior solution, more true to the >> principles of DDD, but not smaller than the dddsample itself. >> >> If we go "EQUIVALENT", then we will not showcase the right thing... >> >> I vote for "RIGHT". >> > > I'm still trying to master the details of DDD. > I'm curious to know what you all think is not "RIGHT" about the dddsample > application (or maybe what is more "RIGHT" about the qi4j approach)...
"RIGHT", as in, an "object" actually has two parts to it: the internal part, such as CargoState, which represents "what the object is" and then the roles/mixin types that represent "what the object does", and where the roles are defined by the contexts and usecases rather than the object itself. In common modeling these two would be mixed into one, which is why the getter/setter problem exists. In Qi4j the raw access to properties is ok, but ONLY within the roles of the object, i.e. they are not exposed to clients of the object. This I think is one of the key insights here! /Rickard _______________________________________________ qi4j-dev mailing list [email protected] http://lists.ops4j.org/mailman/listinfo/qi4j-dev
