Kinetic Object Stream

A Swarm Intelligence framework for creating distributed data pipelines and autonomic neural networks.

It's a framework based on data-centric reactive programming paradigm that dynamically responds to objects in motion through a pipeline of computational actors that can execute concurrently.

Conduct data science experiments, create neural networks, and embrace KOS.

The KOS framework enables dataflow transactions to be expressed as a set of discrete atomic reaction(s) that automatically executes based on one or more desired input(s) it observes from its incoming flow of data stimuli.

Using KOS, you can create a new Persona instance with one or more Reaction functions and feed/pipe a stream of Pulse into the Persona.

One of the most important concept in KOS is that these reactive functions are never explicitly called by other functions as part of a control flow logic. Instead, the reactive functions are invoked automatically by the KOS framework when its input states are eventually satisfied by one or more observed flow of data stimuli. Its operating behavior is very similar to how a runtime garbage collector works tirelessly behind the scenes in a given software instance. You don't need to explicitly tell the garbage collector to clean-up, it simply reacts autonomously based on its environmental observations.

When you embrace KOS you are embracing chaos itself by giving up your programmatic execution flow control logic. Instead of honing your incantation skills and declaring the chain of commands for fulfilling your desires, you've now become an alchemist, continually experimenting with chain reactions and tinkering with stimuli you inject into the system until you get your desired result.

Once you learn how to express and leverage the power of chain reactions, you can create neural networks that evolve based on its operating environment.

NOTE: this documentation is out-of-date and will be updated shortly.

Since KOS encapsulates a number of fundamentally different paradigms for expressing software itself, it is highly advised that first-time users carefully read all of the included documentation below:

• Introduction to KOS
  • Core Concepts
  • Core Entities
  • Programming Paradigms
• Using KOS
  • Getting Started
  • Loading Reactors
  • Sending Stimuli
  • Triggering Reactions
  • CLI Reference
• Clustering KOS
  • Full Stack
  • Hive Mind

The developer's guide to creating new reactors and understanding the KOS library APIs will be coming soon! In the meantime, taking a look at the available reactors bundled with the KOS framework should provide some indirect guidance.

$ npm install kos

For development and running through examples, you can grab the source repo from:

$ git clone https://github.com/corenova/kos

The following Reactor modules are included inside the KOS repository (see /reactor):

Apache 2.0

This software is brought to you by Corenova Technologies. We'd love to hear your feedback. Please feel free to reach me at [email protected] anytime with questions, suggestions, etc.