ITIL needs cybernetics

Dear Axelos, ITIL Needs Cybernetics

In ITIL 2011 – an IT service management (ITSM) best practice framework – there’s five books covering nearly thirty ITSM processes. And as to be expected with any large publication covering a complex sociotechnical system there’s a large, hidden undercurrent of human behavior flowing through it.

In my opinion, the human aspects within ITIL feel like the dark matter of the universe – in that they are obliquely referenced and we can infer that they are there because of the processes that exist to “cope” with human behavior. When ITIL projects go wrong, people are often blamed, as is the technology, but it’s rarely the framework’s fault. And to single out individuals for blame in such a complex system is inhumane – with no mechanism described in ITIL to help people succeed in the ITIL framework. It’s a regrettable oversight, in fact it’s a failure of the framework. However, this issue can be addressed through the use of cybernetics in ITIL.

Oh, the Humanity!

You might have heard of systems thinking, you might have heard of sociotechnology, and if you consider humans and their social behaviors working with technology inside a system, employing amplified feedback loops and continuous learning, and use words like “empathy” like you mean it, then you are thinking about cybernetics.

The essence of human nature is the ability to learn, adjust, and invent as well as having a sense of community and altruism. But human nature can look like Gaudi’s cathedral compared to the Brutalist concrete of ITIL’s repeatable, manufacturing-like process. ITIL has a process-first focus that seeks to enforce uniformity, conformity, and command and control – with human creativity needing to be subservient to these, otherwise it is feared that chaos will ensue.

Human creativity in such systems is blamed for system entropy, such as configuration drift, and configuration drift is blamed for outages. Therefore, ITIL assumes, human creativity causes outages and we are all slaves to fear-driven KPIs, such as mean time between failure (MTBF), which drives a process that in turn drives humans to make no changes at all. No fixes, no upgrades, no new features – which is all bad for business.

What Is Cybernetics?

Compare ITIL’s approach to fields such as cybernetics, the science of communications and automatic control systems in both machines and living things, as explained by Jeff Sussna in his book Design Thinking, and culture movements such as DevOps, which explicitly call out human characteristics such as empathy. At a recent event in London called Operability.IO, the majority of the talks were about human feelings, and using empathy to improve organizational behavior. The failure of humans to understand each other and to get on is characterized by the crevasses between IT departmental silos, and it’s identified as a significant contributory factor to organizational dysfunction and low performance.

These alternative approaches to the anti-human approach of ITIL recognize that failure happens in complex systems. So instead of favoring MTBF, why not focus on improving mean time to recovery (MTTR)? There is also a focus on continuous delivery (CD), which people often mistake for “constantly changing” when in fact all it seeks to do is to increase the success rate of change.

There is also the concept of a “blameless postmortem,” pioneered by John Allspaw, CTO of Etsy – with organizations focused on MTTR, continuous delivery, and blameless postmortems. The latter is described by industry thought leaders, such as John, as: fearless, confident, engaged, and human. As Jeff Sussna says in his book, Designing Delivery:

“Blameless postmortems have an important psychological and organizational effect. They treat engineers as intelligent people trying to do the right thing rather than as untrustworthy, potentially defective, cogs in the machine. By treating team members with respect, it puts another nail in the coffin of the Taylorist industrial approach to employee management. Complex socio-technical systems require human initiative and creative decision making. Blameless postmortems make a critical contribution to sustaining that spirit within IT organizations.”

In my opinion, none of this is visible in ITIL today.

Adding Cybernetics to ITSM

If you meet an ITSM professional, or an industry expert in ITIL, then you will find that they are inquisitive, helpful, experienced, and bright people who want to help their fellow man to achieve good things. They are often the biggest critics of ITIL and constantly discuss how it could be improved upon, though sometimes an ITSM professional might slip into “meat bag blame mode,” accusing people of being the reason that ITIL fails. “You can’t implement ITIL, only its processes!” they shriek.

These ITSM professionals are crucial to the addition of cybernetics to ITIL, and the change of ITIL ownership to Axelos represents an opportunity for both ITIL and the ITSM community to evolve. With cybernetics and DevOps having great momentum in the IT community, the stars might just be aligned.

Here are five practical examples of how cybernetics could evolve ITIL:

  1. Metrics modify behavior: move the focus from MTBF to MTTR in incident management.
  2. Add blameless postmortems to the problem management process and encourage operational transparency and disclosure to all interested parties, especially customers.
  3. Move the service desk from being an edge interface between customers and the service, to the center of the feedback loop – acting as a key step in planning, deciding, analyzing, and acting.
  4. Rethink the design of processes to design for their failure, and how to cope when things go wrong. Introduce “game days” and other cybernetic approaches. This is especially true for security.
  5. Move away from “best practice” and “standards” to a learning culture, one that doesn’t blindly implement dogma (and subsequently criticizes the dogma when it doesn’t work).

That’s my 2 cents, what do you think? 

Image credit


Posted by Steve Chambers

Steve Chambers

Steve Chambers is a recovering IT practitioner and executive with over twenty years experience across the IT industry. From early beginnings in the world of the mainframe at a UK bank, Steve joined the dotcom years working for Loudcloud and working with the UK Cabinet Office on the direct.gov.uk precursor to Digital Government. Steve spent some intensive years at VMware helping customers understand the impact of virtualization on their service management, before kickstarting the converged infrastructure market with VCE and then becoming a Cloud CTO. Steve is using all of that battle experience to help Viewyonder clients get value out of Cloud and DevOps. Follow @ViewYonder for company info, or @stevie_chambers for a personal insight to the industry.