A hybrid command language to merge meaning and reason from cognitively designed languages such as human language and logical machine environment. The language should utilise mathematical functions and reasoning to define the overlap.
Each command must have atleast one of either segments - O L & A
Olacl compiler is the compiler to transform OLACL commands into intermediatory instruction set, referred as goal.
Each compiled goal is executed by the thirdbrain cognitive system - which may or may not result in a knowledge. For example - a goal to observe a sorting program may result in a knowledge of type algorithm.
The resulting knowledge should be persisted as it shall be utilised by the system to decompose goals.
Observation capabilities of thirdbrain are the main driving engine for knowledge accumulation.
Thridbrain observes information streams. As can be expected a single stream of information doesn't fit for purpose. And so, processing a massively parallel streams of information is the core feature of the cognitive system. For example - streams of information from different components of a massive IOT.
The system should be capable of understanding mathematical inference from the observation. It should have the knowledge of all machine learning algorithms and theories to infer relationship across observations. The best possible design to achieve this is a work in progress.
The system should be able to act on input based on the knowledge. The action is the superset of a number of possible solutions to get the desired end results. An action can be anything, for e.g, to synchronise a process, to discover relationships.
It should be considered that to the system itself, observation and learning are only two actions in specific context.