Reference implementation of a consumption layer over the web
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Web-Wide-Matrix POC

Join the chat at

What is this

This project is a reference implementation for an effective consumption layer over the Web, transforming Content Feeds into Organized Spaces. For more information, visit

It is based on several simple idioms:

  1. It defines an environment which is recursively & infinitely scalable, outward & inward.

In particular, this implementation uses the concept of a grid of buildings. Each building can be an atomic container of JSON payload, or it can be composed of floors. Each floor can also be a container of JSON payload, or it can contain a grid of buildings. & so forth.

  1. It allows streams of input & output to the real world.

In particular, this implementation has data-pipes that connect to a bldg floor & either stream into it data items, that turn into buildings, arriving from an external service, or stream out of it each new building, unto an external service.

  1. It contains also actors that move around inside the environment & try to create value by processing data & creating new data. They aren't told how to do anything, but instead try out things & learn by themselves based on how much value they were able to create.

In particular, this implementation has residents moving inside the buildings, that apply actions (external API's) on the JSON payload, to transform the payload & create new buildings. The residents process the data arriving from some input data-pipe, by applying on the JSON payloads API's available on the Internet, that result in either more data inside the payload or new buildings, in the same or different floor. The final buildings that result from their processing are delivered to users via output data-pipe, & the users feedback on these results is used to reinforce the residents learning, so that they improve in creating value.

  1. The process in which developers can affect the behavior of this system is by providing knowledge or training environment that train the residents in achieving some desired value creation. However, once the system is operational the residents may alter their knowledge to adapt to changing user needs or data inputs.

To run:

  • redis: redis-server


  • workers: celery -A mies worker -l info -B --concurrency=4 -Q default

  • residents: celery -A mies worker -l info --concurrency=4 -Q life_events

  • actions: celery -A mies worker -l info --concurrency=4 -Q actions