- Martin Sykes
Enterprise Architecture is changing
Updated: May 1, 2019
“It is not the strongest of the species that survives, nor the most intelligent. It is the one most adaptable to change” Darwin
Change has never happened this fast before, and it may never be this slow again.
In 1965 Gordon Moore, Intel founder, observed that processing power was doubling every 18 months. A law that seems to continue even though I can recall reading articles for the last 10 years telling me that it will be impossible to continue this steady march next year. At the same time as hearing that technology is changing ever faster I observe in organisations a growing need for change management, because change is hard.
It is hard if you don’t want to change. But look at how fast people will change their opinion, their phone, their shopping patterns, when it suits them.
Bill Gates’ made the suggestion that “We overestimate the change that will happen in the next two years, and underestimate the change that will happen in the next ten”. Ten years ago a project I worked on took two years to upgrade 400 servers in 3 data centers. We spent months on the design for the infrastructure to provide identity services, file stores, print queues, internet access and some application servers. Today I can deploy that same capability to a world class standard with a cloud in a matter of days just with a credit card. So much of what I had to do 10 years ago is now built into the service. I no longer get to make detailed customisation choices, but perhaps the end result runs much better for that as it now works to a standard reference architecture proven in millions of implementations.
Many over the age of 40 in IT can remember the days when you might upgrade your sound card. We would set jumpers on the motherboard to define the interrupts to use, or spend hours on the details of a config.sys and autoexec.bat file. When was the last time you did that? Back then these sort of tasks needed a lot of IT knowledge. How much does your average iPhone owner know about how their phone works?
Early adverts on radio were simply the same copy that was printed in newspapers, but read out on air. For decades we saw the growth of TV advertising, then along came the internet and in just one decade we saw advertising change. The wealth of Google is not built on people paying for search results, but on the advertising sold with those results. It is no surprise that the first Chief Digital Officers came from a marketing background. There was nowhere else left for them to go – their whole industry had moved to be digital.
Compared to advertising and many other professions the IT world is relatively young, and within that the IT architecture world is younger still. The initial Zachman framework, from which the discipline formed, was first published as “A Framework for Information Systems Architecture”” by John Zachman in a 1987 article in the IBM Systems journal. But age is a relative term when compared with the pace of change.
Zachman published just 6 years after the first IBM Personal Computer. In 1987 a popular computer was the Atari ST. Based on a 16/32 bit Motorola 68000 CPU, with 512 KB of RAM, a graphical user interface, and the latest in storage technology, the 3.5 inch microfloppy disk.
In the late 90’s organisations had big IT systems. From airline booking to ERP we had complexity. The Mosaic web browser was only released in 1993, and few of the names from the first web boom of 2000 are still with us. Complexity existed in systems that often operated solely within the walls of the enterprise.
Android was unveiled in 2007. By July 2013, the Google Play store had over one million Android applications published, and over 50 billion applications downloaded. At Google I/O in 2014, we heard there were over one billion active monthly Android users, up from 538 million in June 2013. Today we have three competing OS platforms in mobile and enterprise markets with each having user populations exceeding one billion.
In the 1990’s if you had asked a senior executive in an organisation about their enterprise architecture they would have assumed you were talking about the building and facilities management group. That was the period when I first received the “architect” title, and it took quite a while before people started to work out what the role involved and how it related to their own responsibilities.
TOGAF, possibly the most commonly used and cited of the architecture frameworks, started in the 1990s with a focus on technical architecture, but since then has been developed by The Open Group to be an enterprise architecture framework which now has tens of thousands of certified architects.
Many other enterprise architectural methodologies and frameworks have risen and fallen over the last 25 years. Many of the frameworks, taxonomies and methodologies in use in organisations today are still loosely based on the original Zachman and TOGAF structures.
Many of those we have today face a similar set of problems. IT systems have grown in complexity, not because individual solutions have grown, but because a system is now an interconnected set of applications and services spread over infrastructures that extend well beyond the walls of the modern enterprise.
Does this mean the frameworks architects have learned for the last 25 years are flawed or worse, wrong? No. Just as Greek and Roman architecture is still taught in architecture courses for a modern building architect, we need the fundamentals. These are concepts that every architect must understand and help those learning about architecture to “Do the right thing” and "Do things right”.
But just as modern building architects then move on to the latest materials and design technologies we must do the same. The complexity and scope of systems today, with the pace of change, mean we need to adjust some perspectives and understand that the frameworks are necessary but not a complete solution to today’s requirements.
Remember the complexity of the IT world when Zachman and TOGAF were founded. The requirements for Architecture have changed. The systems have grown outside many enterprises. Enterprise Architecture is now too limited in scope!
Leading organisations have recently been extending their frameworks, evolving to include cloud, mobile, big data and social elements. As they continue the evolution to become digital first enterprises the interconnections of data and process must be governed well beyond the boundaries of the enterprise. But when we step outside the enterprise our governance processes may no longer be possible or applicable. Architecture started as a means to manage what in comparison to today was a fairly stable environment, to constrain costs, control complexity and help to achieve planned returns on investment.
The reality is that we no longer have the level of control over the environment that existed when architecture was born, or even for its first 20 years.
Just because we no longer have the same ability to control the environment does not mean the requirements have gone away. They have just become much harder to satisfy.
“We cannot solve our problems with the same thinking we used when we created them.” Einstein
I don’t profess to have foolproof answers to the issues I have been writing about.
But I do know we need to try.
And we need to start now.