NetObjex Blog.

Ecosystem Computing

Created on October 6, 2017

Periodically, I try and take a step back to observe patterns in the way in which information architectures evolve. Over the years, I have witnessed a number of such patterns including:

  • Mainframe computing: A server centric computing model with dumb terminals as client devices.
  • Client-server computing: A form of computing where most of the UI/UX and some of the logic was handled on the client side while the server handled the persistence layer and some of the application logic.
  • Cloud computing: A thin client model using browsers for UI/UX and most application logic on the server side but usually distributed (not a monolith unlike many mainframe systems) among many nodes using a service-oriented architecture (SOA).
  • Mobile computing: Similar to cloud computing with thick client apps which handle UI/UX and some logic and persistence, communicating with a backend set of nodes based on a SOA architecture.

This is by no means an exhaustive list, and there are many variations for each of the above models, as to where and how the UI, logic and persistence layers are sliced up – that what software architects do for a living.

The past year, as a result of my exposure to distributed ledger technologies, I feel we are upon yet another model of computing. Distributed ledgers such as Ethereum enable one to combine logic, database and network (like Ethereum). Not all distributed ledgers are made the same, there are variations in capabilities e.g. consensus algorithms, data structures (e.g. Ethereum uses blockchain, but IOTA uses Directed Acyclic Graph). Regardless of these nuances, distributed ledgers are changing the way in which information architectures are formed within and among organizations.

For a long time now, information architectures were bound by the confines of an organizational unit e.g. enterprise, government, home etc. However, as we leverage distributed ledgers for sharing information, the boundaries of an organizational unit tend to get blurred.

To put a moniker around the information architecture that involves multiple organizational units, I am inclined to calling Ecosystem Computing. In Ecosystem Computing, there can be multiple organizational units participating in a given ecosystem for the purposes of sharing information. Here are a few such ecosystems to help crystallize this concept:

  • Supply Chain Ecosystem: The flow of components from potentially multiple suppliers to an assembly or manufacturing process to culminate in end products (finished goods, value added components).
  • Food Ecosystem: The often used phrase these days is “Farm-to-Fork” to describe all the steps a piece of meat/produce goes through from the time it is cultivated to the point it ends up on your dinner plate.
  • Cloud computing: A thin client model using browsers for UI/UX and most application logic on the server side but usually distributed (not a monolith unlike many mainframe systems) among many nodes using a service-oriented architecture (SOA).
  • Transportation Ecosystem: One that ties that flow of people, vehicles(plane, train, ship and car) and routes.

Each of these ecosystems usually involves multiple entities and with distributed ledgers, information sharing between them is very different from the other computing models because now each entity does not maintain its own data silo and application logic (e.g. through smart contracts) is not contained within a single organizational unit. Why does this matter?

Well, since logic and data are shared, then the concept of owning one’s own data and computing silo sort of disappears. Instead, ecosystem partners will co-own these resources. This is best captured by the diagram below:

The best analogy I can think of is – imagine the railroad is operated by A, there are multiple train operators B,C,D and there are users E, F,G. The train operators need to pay a fee to A to run their trains, and in turn charge users E,F,and G for services rendered (transporting people, goods etc).

In ecosystem computing, the ticket to participating in that ecosystem is a token. A token is a cryptocurrency which will most likely be the unit of value by which all usage of services rendered will be metered and paid for. So, ecosystem computing is here and is already introducing new and interesting monetization models for computing resources just like SaaS disrupted the older packaged software licensing model of the prior generation.

Related Articles

About Us

Created on October 6, 2017