Semantic blogging combines blogs with the Semantic Web for improved metadata. Our analysis of six semblogging approaches and systems reveals that all are RDF-based, suffer from not using Published Subjects as proxies for subjects, and do not employ semantic relations motivated by knowledge organization. In contrast, we introduce and discuss semblog-tm, a prototype of a Topic Maps- and PSI-based semblogging system with a well-motivated set of semantic relations. Its basic requirements have been derived from the three main use cases: managing lightweight ontologies, attaching semantics to blog entries, and providing aggregated semblogging data as semantic knowledge services to other semantically-enabled systems. The four main system components (blogging, topic map, PSI, semantic knowledge services), and its capabilities are sketched. The web application semblog-tm is implemented as a plugin for the blojsom blogging engine, using Java and Tomcat with Velocity templates. Topic map operations are realized via TMAPI with TM4J or OKS, and Axis is employed for web services.