eRRA - The eReading Reference Architecture

Previous page : The eReading Market
The basic assumption behind this document is that, in the following decades, eReaders will become mainstream. Therefore many applications will be built . A reference architecture can be a great way to speed up the design of such an “eReading Application”.

An architecture defines the elements of a system and their relationship with each other.

A reference architecture is an architecture one would like to refer to. It's a template. A starting point for thinking about a new application, just like as if Building Architects had common plans they use to start designing a construction.

The eReading Reference is based around three important ideas.

  1. Four pillars are necessary to build a complete eReading system: the eReader, the Web, the Desktop and the Entreprise.
  2. Each pillar is described according to Archimate's structure which counts 9 "categories". The complete architecture classifies everything into 45=36+9 categories. 36 categories for the 4 tiers and 9 categories for common stuff.
  3. eReading contents has a life-cycle : the eReading Content Lifecycle.

The 4 tiers of the Architecture

When you hold an eReader in your hands (an iPad, an iPhone, an Android phone or tablet), you only see the visible part of the eReading Iceberg. Several systems had to be used in order to present you the information:
  • a Desktop computer was used to create the content
  • an Entreprise server was probably used to manage this content (store the images, the text, the videos...)
  • Web servers where used to publish this content
  • finally the eReader was used to display the content so you can read it
To 4 tiers are described in detail in the section called The 4 Tiers of the eReading Architecture.

The 9 archimate "categories"

The eReading Reference Architecture is based on the Archimate meta-model (see http://www.archimate.org). You don't need to worry about Archimate if you don't want to. It's just a framework to organize the architecture. The goal is to better identify the nature of an architecture element : is it an Actor ? Is it a Process ? Is it Software ? Is it a device ?

Three layers are important in Archimate : the Business, the Applications and the Technology. Into each layer, one will look at the Information, the Behavior and the Structure of the System.

  Information Behavior Structure
 Business Layer
 12
3
 Application Layer
 45
6
 Technology Layer
 78
9

If you take a close look at the Table of contents you will find the 9 categories for each of the 4 tiers.

In a first phase, the Reference Architecture has been developed using UML (http://www.uml.org) and the BOUML UML toolbox (http://sourceforge.net/projects/bouml/). You can see the Archimate Meta-Model designed in BOUML in Appendix - The Archimate Meta-Model in BOUML.

The collaboration between the tiers results from the Interaction of different processes. This is the eReading Content Lifecycle.

Next page : The eReading Contents Lifecycle


Comments