Design: Business Design Induction/Deduction

hex-business-design-01

This is my latest incarnation of the Business Design Process.  Induction (Brainstorming–generation of ideas) is Counter-Clockwise.  Deduction (Refinement–elimination of ideas) is Clockwise.

Below is the Intelligence Architecture:

hex-business-design-03

Here is the Media Architecture:

hex-business-design-04

This is the Data Architecture for this model.  Note that all values are accepted even if they are wrong:

hex-business-design-05

Below is the Network Architecture of this model.  Note that the values are unique (nodes) and they are sequential (edges):

hex-business-design-06

Here is the Text Architecture:

hex-business-design-08

Here is the Numeric Architecture:

hex-business-design-07

Here is the Octonion Architecture:

hex-business-design-octonion

The Zen of Systems and Networks

zencircle01

My own work with Enterprise Frameworks and Networks has led me to come up with the following table.  It describes the Nodes and Links in a Complete System Network.  I am saying that the Nodes representing Goals, People, Time, Locations, Code, Data, Qualities and Quantities can all be represented as Scale-free Networks and that each of these Node Networks require only one datatype.  I am also saying that there are only three types of links in networks: recursive links within a set, multiple links between sets, single links between sets.  I know of no case where this has been attempted in the manner I am attempting to represent it.

systemnetworkslinks1

If you have been following my blog you are aware that I have been struggling for a long time to come up with a framework and a clean terminological set to describe systems.  I think I have come one step closer to that goal today.  The table above describes a Fact composed of eight Nodes (first white row illustrating entities) and the Links (last three white rows illustrating recursive, multiple and singular relationships) for each of the System Networks (Interrogative columns).  One of the interesting aspects of this System Network Model is every Fact is composed of a Unique Set of all eight Nodes.  However, all the Nodes in one Fact do not have to have Links to all the Nodes in another Fact.  Each Node within a Fact is independent regarding its Links.  Therefore you have a single set of System Facts with each Fact containing a single set of Interrogative Nodes each connected by their respective Link Networks.

I have recently been writing with the intent to challenge centrism on any one of these networks and advocate a more integrated view. I still remember dealing with data centrism, event centrism, user centrism, goal centrism, program centrism and schedule centrism over the course of my career. All of them have a role to play. My insight into all of these Nouns being Linked by Verbs in only three ways required me to look at all of the Enterprise Architectures and disengtangle the Nouns, Links and Verbs from the reasoning and representations that extend back beyond computing itself.

The Data Model below is a hybrid of Relational Models and Dimensional Models.  I call this an Associational Model.  It is using Relational Architecture to represent it.  However, I think that an alternate Entity-Attribute-Value (EAV) architecture called the Associative Model of Data would be better suited to the task.  I am using relational representation as I am still trying to communicate with a community only familiar with Relational technology.

The first thing to note about this model is Links are represented by Associations.  Associations link two Nouns using a Verb.  What is interesting about this model is every Verb, Association, Noun and Fact is unique.  The vertical connections are Many to Many relationships which allow two vertically adjacent Verbs, Associations or Nouns to have multiple unique relationships between each other.  What this means is there are no integrity problems (duplicate values) as the system network would enforce uniqueness.

amdschema031

The premise of this model is that the Nodes are not dimensions at all.  I am rejecting the traditional concept of dimensionality instead I am saying that there are three dimensions of Links: recursive, multiple and singular.  All we perceive are Facts, Nodes and the Links between them.

So you could come away with the following Zen koan:

entity without entity,

source without source,

path without path,

target without target,

size without size,

dimension without dimension.

Systema: Seven Hats, Seven Links

watch-parts1

Parable of the Watchmakers

There once were two watchmakers, named Hora and Tempus, who made very fine watches. The phones in their workshops rang frequently; new customers were constantly calling them. However, Hora prospered while Tempus became poorer and poorer. In the end, Tempus lost his shop. What was the reason behind this?

The watches consisted of about 1000 parts each. The watches that Tempus made were designed such that, when he had to put down a partly assembled watch (for instance, to answer the phone), it immediately fell into pieces and had to be reassembled from the basic elements.

Hora had designed his watches so that he could put together subassemblies of about ten components each. Ten of these subassemblies could be put together to make a larger sub- assembly. Finally, ten of the larger subassemblies constituted the whole watch. Each subassembly could be put down without falling apart.

sevenhats2.jpg

For the longest time I have been playing with interrogatives and associations.  Now, I think I finally have a complete representation and taxonomy.

Abstractly, it looks like the following:

enterpriseabstract3

Concretely, it appears as follows:

enterpriseseven5

As I mentioned in my earlier post, I was not satisfied with a six interrogative, four association model.  Consequently, I worked to resolve this and came up with the table above with the interrogative columns (seven hats) and the associative rows (seven coats).  I also came up with the data model below:

enterprisefact1

My hypothesis is, used correctly, the above data model can address all relational/dimensional requirements.

Related Posts:

Framework for a Real Enterprise

It was Peter Drucker who revealed undeniably that business was a science that could lead to predictable results.  The way he did so was by collecting and systematizing all the knowledge he could gather on the subject and then testing hypotheses.  After much deliberation on the science of systems and the science of business.  I present the Physics Framework above and the Enterprise Framework below.  As one physics Nobel laureate said, “If you aren’t doing physics, you’re stamp collecting!”

I am working to refine my framework table for a lay audience. It is a vocabulary for a business system. Like the Linnean system, by using the intersection of the row and column (two terms) I can identify any operation of the system. Still needs work, but its getting there.

It is based on an associative (node and link) architecture not a relational (table and relationships) architecture.

At first glance this might be regarded as a Zachman Framework.  The columns by convention are called focuses.  The rows called perspectives.  The interrogatives make up the column header.  John Zachman offered some poorly chosen row headers which I’ve replaced.  There are two major differences.  First, it requires an additional focus as part of the enterprise, the Market which is measured in potential profit.  It’s time for the academics and bureaucrats to stop turning up their noses to the source of their existence:  a market that will pay in currency to fatten their budgets.  Second, REVISE, the nodes, are something obvious to Einstein; RELATE, the links, something obvious to Drucker (remember the links are verbs); REPORT, the node and link attributes, should be obvious to Thomas Jefferson; RECORD, the databases, to Carnegie; REGARD, the datatypes, to Turing; REPOSE, the ordinality, which remembers whats related to what, REVEAL, the cardinality, full of exceptions to the enterprise.

Systema Framework Datatypes

A new insight that has come to me in working with the Systema Framework is the realization that datatypes are not chaotic or random.  Datatypes are specific to the column they are in.

  1. Logic is restricted to Text attributing the interrogative
  2. Organic is restricted to Integers attributing the cardinality
  3. Heuristic is restricted to Boolean attributing the paths or sets
  4. Pragmatic is restricted to Decimal attributing the metric
  5. Chronic is restricted to Datetime attributing the event
  6. Cosmic is restricted to Spatial attributing the node

The interesting thing is the framework is two dimensional.  This means that each of the six interrogatives is provided with each of the six datatypes in a structured fashion.

When you give this thought in this disciplined fashion, it becomes obvious that the Zachman Framework and the Systema Framework are inadequate to explain the complexity of datatypes we experience.  I will provide a solution in my next post.