Phenomena are characteristics of real world objects, the value for which may be estimated by application of some procedure in an observation. Note that the phenomena can be “real world” phenomena but also phenomena in "hypothetical worlds" like in simulations.

Whenever we record or process data of a phenomenon in order to describe the phenomenon, this data will only be an abstraction of the "real world". E.g. if the phenomenon we are interested in is the temperature of some fluid we will usually use some kind of thermometer to measure this temperature. Even if the thermometer was indefinitely precise it would still only measure the temperature of the fluid only at one point and would not represent the real world phenomenon exactly. The abstraction of a real world phenomenon that we deal with in Redbex is done using features.

Different kinds of phenomena result in different kinds of features these kinds of features are called feature types. The feature type defines its set of properties that are used to describe features of this type.

An Observation is the act of observing a phenomenon, with the goal of producing an estimate of the value of the phenomenon at a specific point in time. An observation is a specialized event whose result is called an observation result. The observation result is a set of properties that are defined by the feature type of the feature the observation is done fore. Note that we use the term observation for both, the act of observation, and the observation result. Since Redbex deals with observation results the term observation is usually unambiguous.

An entity capable of observing a phenomenon and returning an observed value is called a sensor. We speak of a sensor as a source that produces a value within a well-defined value space of an observed phenomenon. Note that a sensor can be some device as well as a human being.

To observe different kinds of phenomena different kinds of sensors are used, these different kinds are called sensor types. A sensor type defines a set of properties that describe sensor characteristics which do not change over time, a set of properties that describe characteristic of a sensor that is valid for specific time spans and a set of properties that describe characteristics of a sensor that change which each observation.

These very abstract definitions of phenomena, features, observations and sensors are key concepts within Redbex. The generic implementation of Redbex based on these abstract definitions is the key to the flexibility and broad applicability of Redbex.