DevOps in a Pandemic

Programmer productivity during lockdown is an easy story to tell: Instead of water-cooler talk and keyboard work in the office, DevOps practitioners Zoom and Slack their way to the same keyboard work from home. Right?

Wrong! There’s so much more to the DevOps story at this point in history.

Working through a crisis

Conventional wisdom says that work in 2020 is the same as in 2019, but prefixed by “WFH” (working from home). This is wrong, in several regards.

Fundamentally, all of us in most of the industrialized world are working through a crisis. Disruption has lasted weeks at a minimum essentially everywhere, and the US in particular seems to be creating months or even years of risk and loss. These are far from the best of times, and the default assumption should be that every coworker is dealing with family tensions, busted budgets, medical emergencies, and the fatigue that uncertainty and threat bring.

For some of your team members, WFH is business as usual, but others might be poorly equipped to use their homes as workspaces, even in the best of times. Different workers have different home environments. Some are easy fits for WFH; for others, WFH is possible but an ongoing strain.

Managers are also variable. Some are well-equipped for WFH; they know how to encourage their teams to such an extent that group productivity actually rises during the pandemic. Others, who were adequate leaders during normal, in-office operations, are so unfamiliar with remote ways of working that they end up degrading not just their own results, but those of everyone who reports to them.

Being successful from home

Even if we ignore all other confounding complications, WFH is different in many of its fundamentals. Respect for those differences is a prerequisite to success.

Plenty of guides to making the most of WFH for yourself, your employer, and your coworkers are already freely available, from Harvard Business Review to Vogue. The main thing to keep in mind is that nothing is automatic. Individual situations and team dynamics interact in complex patterns with the bare facts of WFH. It’s a great time to be explicit and deliberate about taking care of each other, and to think in terms other than “how we normally do this.”

Is it a priority, for example, that the team still meet a particular release deadline? They’re adult enough to hear that message; it becomes far more credible and powerful, though, if the organization simultaneously erases administrative overhead, invests in tools or otherwise demonstrates its grasp of how priorities work. Make WFH an opportunity for fundamental improvement, not just another occasion where sacrifices of front-line workers buffer the organization from mishap.

Enforcing cybersecurity, remotely

What about DevOps practitioners in particular? What are specific issues for us?

One of the most important is security. Bad actors are industriously looking for novel vulnerabilities, and the basics of migration to WFH have overwhelmed many IT staffs. DevOps workers frequently are serving as their own support.

How can an organization respond? First, make it easy for programmers to do the right thing. If a family member accidentally publishes a work password from a shared machine, it’s not a good thing, of course. When something like that happens, though, we want the employee to report it swiftly and help with countermeasures. Turn the pandemic into a time to enhance transparency, as a counter to the threats and difficulties of WFH.

DevOps skills even have the potential to serve other teams. The analysis and automation that DevOps professionals specialize in might be able to help overburdened IT and security teammates with specific chores.

Keep in mind that security is always an exercise in defense-in-depth. A good plan encrypts data, hardens WFH workstations, locks down applications, restricts access, and so on, in a balanced way. No single technology or rule answers all the needs of security.

Initiatives

The pandemic is a great time to invest in more enlightened policies. Older rules that assumed what happened at keyboards in offices are the only things that mattered are ripe for update. Now is a time to unify and consolidate policies so they make sense across different workplaces, including WFH.

Make the most of human talent by bringing silos together for more consistent operational security, monitoring and response. While it takes effort to build functional consoles or dashboards, it pays off quickly in enabling consistent and swift incident management.

Complementary organizational projects are equally timely. Now’s the time for organizations to switch from a command structure to stronger goal orientation, and from reaction to systematic improvement. Decision-makers themselves are likely overwhelmed, and anything that enhances the autonomy of DevOps programmers is likely to pay off. The pandemic strains all sorts of systems, and there’ll be needs to deal with illness and other emergencies apart from infection specifically by COVID-19. It’s a great time to work on disaster recovery and business continuity goals.

Customers are living their own pandemic dramas too, of course. Their expectations about time and space are changing. They know, for instance, that on-site service is increasingly unlikely; at the same time, they might put a new premium on round-the-clock capabilities to help with their own WFH staff. Internationalization, localization and accessibility also are increasingly important for customers who are now shifting away from the geographic borders and barriers that governed their operations in the past.

Change and failure are abounding during this pandemic. Improvement is also possible, though. DevOps can be at its best if we put into practice some core professional values:

• Agile response
• Continuous learning
• Fierce focus on goals
• Peer review
• Automation

If we do this right, our DevOps teams can emerge as new centers of strength for our organizations.