In game studies, ontology is the study of the nature of games: their mode of being or existence, and of variation within their domain. However, this is vastly complicated by the fact that game studies is constituted of a great variety of methodological and disciplinary approaches, many of which do not study the same type of phenomenon (e.g. an ethnographic vs. a technological approach). So the preliminary steps of any game ontol-ogy must be to first establish a meta-ontology (or more precisely, a meta-game-ontol-ogy), and then place itself within it. Since a comprehensive list of approaches (and due discussions of these) would demand much more space than can be allocated here, this approach will be fairly general, with many omissions and simplifications. The term ontology may refer to the general study of being and existence, or a particular theory of being and existence, but also to the meaning used in computer science, that of a formal mapping of an empirical domain (e.g., a railroad system), and the construction and use of such descriptions in implementing simulation-or control-software that accurately models behaviors, objects, and their relations within this domain. Game ontologies can have similar motivations: they can be highly specific formal models of the design space of games, or they can try to answer the general questions: What are games? What do they consist of? Where are they in relation to similar phenomena? Thus, we have at least two different types of game ontologies: (1) Formal or descriptive ontologies, asking what are the functional characteristics and components of game objects, and the relations between them; and (2) existential ontologies asking what are games and what kind of existence does a game have. Both types of ontologies presume that games exist, but neither is dependent on a formal definition of the concept of game in order to be meaningful. So the question of whether it is possible to define the category of games formally, introduced and answered in the negative by Wittgenstein (1953) and challenged (Suits, 1978) but never refuted, need not concern us here. "Games," like "texts" and "planets," is a historical term and not a scientific one, and trying to change it into a theoretical term would probably do more harm than good, were it to succeed. As Wittgenstein pointed out, we can still talk about games successfully without a definition, and, let me add, we can still describe games through formal models. The only risk is the possibility that we also will describe things that probably are not games, but this overproductivity matters not as long as the models convincingly describe the phenomena people call games. The most basic ontological concern regarding games is whether the word refers to an object or a process. Games are both object and process (a combination of states not dissimilar to the duality of language: langue/parole, paradigm/syntagm etc.), but the phrase "a game" will refer to either one or the other, not both. In most contexts, "I