Now we have DevOps , not quite Maxwell’s laws of Electromagnetism

Joining things up = progress: a connection between Physics and DevOps?

Next Story

Say “Sayonara” to Cost Centers,

It sometimes seems IT is just one endless stream of brand new things; service management tempers that a little. Service is an old concept, one that has been delivering customer outcomes for centuries. New things are supposed to be better than old ones, seeing similarities to the past can help us avoid repeating mistakes and deliver better. We use our backgrounds to help make sense of our situation and, in trying to set out my vision of things, I have fallen back on my university physics. But where does DevOps fit into this picture?

Scientists used to seek rules for little bite-sized, small enough to study, parts of how the world seemed to work, they developed rules to explain:

  • How stones move when thrown
  • What happened when things burned
  • How magnets attract iron filings
  • The way light moves
  • How waves travel.

A perceptive Scottish scientist called James Clerk Maxwell in the 18th Century turned this on its head. He looked at three different areas: electricity, magnetism and optics: others saw them as separate things, he showed they were all just one, starting a fashion for fitting everything together that has since led us to the Higgs boson.

Why am I writing this for a service management audience? Because it seems that IT is sorely in need of its own Maxwell.

From Computers to IT to complexity

Originally computing was just computing and some people understood it all. Then Information Technology evolved and moved away from being a craft[1] to being an industry. Instead of a few talented people being creative, we needed lots of people being consistent. IT got more boring but it also got reliable.

Industries need rules and processes and they duly appeared in IT, we got methodologies and frameworks. But they were approaches to doing parts of IT, ranging from simple, sensible rules to sophisticated guidance. Following on the success of approaches such as SSADM in IT development, came the ITIL guidance in 1989, intended to deliver for Operations what SSADM had done for application development.

IT methods and frameworks have proliferated since then and we have lots of them: ITIL, COBIT, TOGAF, CMMI and many more.

With SSADM and ITIL, we had an approach for specifying and building software (development) and another for running it (operations). That made sense, in a pre-Maxwell kind of way.

Now we have ‘DevOps’, not quite Maxwell’s laws of Electromagnetism, but a logical enough search for one structure combining development and operations. Logical because there’s no need to develop if you aren’t going to operate, no need to operate if you don’t develop.

Are we learning from the past or repeating mistakes?

But, are we in the 21st Century, repeating a 19th century attitude? Making one connection is not enough; we still have a long way to go, even on the development and operations marriage. Devops is a start but seems to be driven by the development side. Successful relationships need a vision that incorporates both parties’ strengths, not domination of either side by the other.

Between devops, ITIL V3 strategy and design and a range of other good stuff we have a basis for integrated guidance for the first time in IT.

But, assuming this happened, it would still be shooting at a partial solution. Looking at IT as if were separate from the rest of the organisation.

Seeing electromagnetism as a whole was a serious improvement over treating electricity, magnetism and optics as separate disciplines. We see how changing one kind of thing affects other things, and lets one lot of thinking and calculation predict and control a range of important purposes and outcomes. Similarly a joined up approach to IT, breaking down the invented walls between development and operations will give us efficiencies, economies and greater effectiveness over the eventual delivery of value.

Today we know electromagnetism is itself just a part of a broader integrated set of concepts. Similarly, ITSM is an integral part of bigger things. Self-evidently, it’s part of IT and of more general Service Management. Managing that integrated whole has to be more sensible than letting aspects consider themselves self-contained.

Unfortunately, the expansion approaches we do see have a two-stage approach:

  1. Internal perfection. Focus on one little area and treat it as something that can exist and improve in its own.
  2. Magnanimous domination. The need for other aspects to be integrated is seen, but the perspective is of colonisation. For example: ITIL’s diversification to subsume elements of development and strategy or the development community presuming their attitudes can overrule the need for operational processes.

Of course, we absolutely still need the detail that comes from a tight focus. For example, Network management needs to be really good at managing networks if the whole service is to work, developers need to be smart and adept at building applications or the value can never be provided. But it needs to be done in context.

Gluing our future together?

It is about gluing all the pieces together to build and maintain a service, not just an application. From the IT side of ITSM it means linking the ITIL/COBIT/ISO20K etc frameworks in with development approaches, TOGAF, PRINCE and all the rest too. IT4IT has set out to assemble the range of IT frameworks into some kind of cohesion. We wait with interest to see how successful and popular it becomes. But even that might give us, at best, electromagnetism levels of integration. That won’t give integration with the development, transition, operation and improvement of business services, of which the IT is a fundamental part. Too many frameworks we see still expect IT (and presumably other supporting services) to think and exist separately from the overall entity.

As one minor example, in most companies, finance is a part of any major project from the beginning and so we are used to cross border involvement. And we do see some mention of finance in IT frameworks but too often it is building additional IT-focused finance not integrating across the boundaries. When all the relevant parts of the organisation get to work together (and that includes IT) – then maybe we will be getting somewhere. I am not holding my breath, but we can still be hopeful.

[1] Except in the movies. In the movies and on TV you will still find, in every crime fighting team, a craft oriented expert who, simply by their being a ‘computer genius’ can hack through any security, make any computer do their absolute bidding and conceive, write and implement revolutionary software on a pocket calculator within 5 minutes.

The following two tabs change content below.
In 23 years working for the UK government, Ivor Macfarlane moved from forestry to ITSM via prison, stores and training. He has worked as a Service Management trainer, consultant and writer since going freelance in 1999 and then after a 7 year spell with IBM he is now independent again working through MacfPartners to deliver training and consultancy to customers. He was an author for ITIL (versions 1,2 & 3), ISO20000 and ITSM library and an ITIL examiner since 1991. He is well known at ITSM events having presented at many around the world (39 countries so far) and is an active contributor to social media.