The Semantic Web has become reality over the past couple of years. While certain practical topics—such as interoperability, etc.—have at least partially been addressed, scalability and expressivity issues regarding the utilisation of multimedia metadata on the Semantic Web are still widely neglected. However, existing Web (2.0) applications handling millions of multimedia assets are starting to take advantage of Semantic Web technologies. This work contributes to design decisions regarding scalable and smart multimedia applications on the Semantic Web. Based on an analysis of practical issues—stemming from diverse projects and activities the author has participated in over the past four years—three areas have been identified, namely (i) performance and scalability issues on the data access level, (ii) the effective and efficient representation of multimedia content descriptions, and (iii) the deployment of multimedia metadata on the Semantic Web. The three research areas have as its common base the trade-off between expressivity and scalability. We present our findings regarding scalable, yet expressive Semantic Web multimedia applications in a number of practical settings and discuss future directions, such as interlinking multimedia.
