Summary: This paper discusses issues that arose in an effort to transpose the semantic and methodological value of HRT-HOOD into a carefully-crafted selection of the UML meta-model. The main goal of the project was to prevent the massive advent of the UML from obliterating the unique strengths of the HRT specialisation of HOOD for the support of real-time design. Avoidance of closed conservatism also afforded the project the opportunity to reflect on aspects of object-oriented design that may arguably benefit the resulting HRT-UML method.