Reliving my past

Over 25 years ago I suggested to a range of business users, designers, engineers, surveyors, and planners - that paper based maps, plans and designs or specialised dedicated models suited to a single audience or purpose - were not a good way to bring all the information to together (and allow it to be integrated, maintained, analysed, accessed by a wide range of different parties etc.). In those days I focused on technical computing (mapping, CAD etc.).

In those days I was talking to people who had spent a lifetime learning how to create truely beautiful watercolours that the scrapping little draws that came from that generation of CAD were the way forward.

The challenge was to get different set of people (professions) each of whom sees things from the own perspective yo see the overall picture and what is required to suit the needs to the entire constituency i.e. that if all information was made available using the same framework e.g. co-ordinate system - we could all see what was there, why it was there, what it related to etc. Design work could be migrated to as-built in a fairly painless, predictable and straightforward manner.

Conceptually now - that battle at least is essentially won (though aberrations remain).

Now I am having exactly the same discussion with people seeking to implement complex IT infrastructures (e.g. NSDI) i.e. business users, designers, engineers and planners. In case it is the meta information that is on paper i.e. the maps, plans and designs that sit and describe the: functions, the data, the roles, assets, networks of related elements etc.

Now we have a different set of people each of whom sees things from the own isolated perspective and fail to see the overall picture and what is required to suit the needs to the entire constituency.

NASA Deputy CIO says EA really help control of IT-related costs

NASA deputy CIO mentioned that implementing EA had really helped his agency develop a mature IT organization and control IT-related costs.

Is mention in open paragraph of this


http://www.computerworld.com/s/article/print/9137306/Building_your_enterprise_architecture_program_the_right_way?taxonomyName=Enterprise+Applications&taxonomyId=87

I had to laugh at what followed.

It started with

"I had to agree, " [once senses the author is reluctant to agree to so a prosaic goal as cost reduction]

It continued with

"...but the topic of EA is far more complex ... far more difficult to properly implement than most IT professionals realize" [whereas in fact it properly implement EA initiatives can be implemented quite simply and progressively delivery value]

It then proudly proclaimed

"...there are four main areas of EA, covering business processes, data, applications and technology. ..." [seemly forget about the bulk of the actual enterprise e.g. it goals and constraints, its products and organisation etc.]

Yes these things are important:
- capabilities, functions and processes
- business decisions/questions, information and data
- applications and services
- technologies

But the facts are that costs are directly associated with products and services (i.e. applications, services, technologies, standards and projects that change them and the organistions that support them) and only indirectly related to BP and data.

That income is related to the companies: products/services.

So if you don't understand the where the goes and where the money comes from all the knowledge of what lies in the middle is hard to apply.

Contemporary solutions IT strategic in many cases provide out of the box solutions for many of these EA issues that deliver value quickly.

Sooner or latter these EA people who saying everything is so hard and so complex (and reference obscure abstractions and methods) are going to need to realise that these thing have not worked.