Abstract
The UML sits at an architectural crossroad. Will UML 2.0 resolve the problems of UML 1.x or will it succumb to the dreaded second-language syndrome?