Interoperability is a qualitative property of computing infrastructures that denotes the ability of sending and receiving systems to exchange and properly interpret information objects across system boundaries. Since this property is not given by default, the interoperability problem and the representation of semantics have been an active research topic for approximately four decades. Early database models such as the Relational Model used schemas to express semantics and implicitly aimed at achieving interoperability by providing programming independence of data storage and access. Thereafter the Entity Relationship Model was introduced providing the basic building blocks of modeling real-world semantics. With the advent of distributed and object-oriented databases, interoperability became an obvious need and an explicit research topic. After a number of intermediate steps such as hypertext and (multimedia) document models, the notions of semantics and interoperability became what they have been over the last ten years in the context of the World Wide Web. With this article we contribute a retrospective on semantics and interoperability research as applied in major areas of computer science. It gives domain experts and newcomers an overview of existing interoperability techniques and points out future research directions.

Top