There is rapidly growing momentum for web enabled agents that reason about and dynamically integrate the appropriate knowledge and services at run-time. The World Wide Web Consortium and the DARPA Agent Markup Language (DAML) program have been actively involved in furthering this trend. The dynamic integration of knowledge and services depends on the existence of explicit declarative semantic models (ontologies). DAML is an emerging language for specifying machine-readable ontologies on the web. DAML was designed to support tractable reasoning. We have been developing tools for developing ontologies in the Unified Modeling Language (UML) and generating DAML. This allows the many mature UML tools, models and expertise to be applied to knowledge representation systems, not only for visualizing complex ontologies but also for managing the ontology development process. Furthermore, UML has many features, such as profiles, global modularity and extension mechanisms that have yet to be considered in DAML. Our paper identifies the similarities and differences (with examples) between UML and DAML. To reconcile these differences, we propose a modest extension to the UML infrastructure for one of the most problematic differences. This is the DAML concept of property which is a first-class modeling element in DAML, while UML associations are not. For example, a DAML property can have more than one domain class. Our proposal is backward-compatible with existing UML models while enhancing its visibility for ontology modeling. While we have focused on DAML in our research and development activities, the same issues apply to many of the knowledge representation languages. This is especially the case for semantic network and concept graph approaches to knowledge representations.