The most adaptable of our three characters. The ability of humans to learn new technologies and adapt to new situations is simply amazing. The importance of this is often overlooked in a world dazzled by new technology. But there is an unfortunate side effect of this adaptability. It can cover up weaknesses in process and technology.
For an organization to make meaningful strides on the automation journey, it will require people to drive the transformation of process and technology. For this to happen, a culture is needed that allows people to thrive.
A few things to consider:
- Innovation comes from people. Whether an innovative idea originates inside an organization or outside, it must be discovered, tried, evaluated, and if useful, adopted. It is important that organizations foster an environment that encourages experimentation with innovative ideas.
- Some of these experiments will result in failure. Effort and sometimes money will be wasted, but this is necessary. Organizations need to foster an environment in which trying and failing is acceptable, and the philosophy of fail fast is promoted.
- People will make mistakes, it is inevitable. Some of these mistakes could have a significant negative impact. It is better to treat these events as learning opportunities, than as witch hunts. Organizations should adopt the concept of the blameless postmortem.
As an organization moves toward higher levels in the autonomous framework, new skills are required. These new skills will vary greatly, by organization, by role and by level. To transform their skillsets, people need time and support.
It is impossible to list all the new skills required, but a few examples are:
- Proficiency working with a version control system (git) and source code control tools (GitHub, GitLab, Bitbucket, and so on), because Level 2 expects machine-readable definition files to be under version control.
- Proficiency working with the RESTful API, because it provides a common interface for automating individual components of a system.
- Experience with tools that use a declarative approach to defining infrastructure, because Level 2 expects this approach.
- Experience with one or more of these tools (Ansible, Terraform, Puppet, Chef, Pulumi, and so on), because Level 2 typically means the use of CCA tools.
- Experience with container platforms (such as Kubernetes and Docker) because many times, new architectural patterns (Cloud Native, Microservices, and so on) favor containers over virtual machines.
- A solid understanding of DevOps principles, because a new operational pattern around Infrastructure as Code emerged a while back.
- A general understanding of analytics, because Levels 3-5 expect analytics to play a significant role.
- A working knowledge of machine learning model training, because Levels 3-5 expect machine learning to play an increasing role in the development of models.
- A general understanding of data management techniques is desirable, because Levels 3-5 are increasingly dependent on data.
- A general understanding of MLOps and DataOps is desirable, because new operational patterns have developed around machine learning and the data needed to drive it.
As people navigate the process of transformation during the automation journey, it is increasingly important that communication practices also evolve.
A few things to consider:
- New communication and collaboration tools may be necessary to improve the information workflows, especially in the face of an increasing amount of remote work. Don’t assume that the current tools will be sufficient.
- With the above statement in mind, as new data sources and ways of instrumenting automation are added, it is crucial that these elements are not the ones driving change in communication and collaboration tools. Technology should be able to integrate into existing information workflows.
- When changes are happening to process and technology, it is important that these changes are effectively communicated to people.
- Until an organization reaches the upper levels of the framework, human operators will be needed for day-to-day operations. People need to understand how the above changes will affect them.
- Proposed changes should be discussed with people early on and buy in should be sorted out.