Ending #Failoverconf is @jpaulreed saying hi to his mom and maybe talking about Resilience Engineering

Resilience gets convoluted because there’s a lot of “R” words that tend to get intermixed and used interchangably and that gets confusing. “Also if you say ‘resiliency’ you’re a bad person and you should feel bad”
@jpaulreed #Failoverconf

[ok last part is not a direct quote]

First Robustness: It’s the ability to absorb distrubances. And they’re really well modeled. And we’re solving for a specific situation. It’s Known-Knowns.
@jpaulreed #Failoverconf

Reliability - These are patterns and choices that support consistent delivery. Like having multiple traffic lights for a single part of an intersection.

@jpaulreed #Failoverconf

Rebound - getting back to normal quickly. Like a CI/CD pipeline makes delivering changes and rollbacks easier. Lets you get back to normal easier. But it has to be in place BEFORE the disturbance and needs previous experience. @jpaulreed #Failoverconf

“Resilience isn’t something you have, it’s something you do.” —apparently everyone who works in Resilience Engineering.

@jpaulreed #Failoverconf

COVID-19 is creating some really interesting cases here. Lots of systems failing and stretching.

@jpaulreed #Failoverconf

This brings up Adaptive Capacity as another corner of resilience.

Unique; small, but effective; quickly modified interventions with appropriate signaling mechanisms indicate a high adaptive capacity for a system.

@jpaulreed #Failoverconf

Signals are important, we have to know how people are actually adapting, are they doing well? Are we forcing them to do something that’s not working? Should we start to go back to normal? We need those signals. @jpaulreed #Failoverconf

Adaptive Capacity is about long term support of change. Organizations that are good at deploying quickly, changing quickly indicate a high adaptive capacity, they’re used to consistently dealing with change.

@jpaulreed #Failoverconf

“During this once-in-a-century pandemic, do we stop deploying?”

Interesting, but the wrong question to ask. Right questions:

“Where is the idea coming from (to stop deploying) and where is pressure being applied to do so?”

@jpaulreed #Failoverconf

A change board (blunt end) saying we should stop deploying is very different than a product team (sharp end) indicate that they feel unsafe making changes given the effects of COVID.

@jpaulreed #Failoverconf

Cognitive Load is a real concern in the current circumstance. How do we deal with that impact?

-Shed Load
-Trade off thoroughness
-shift work in time
-recruit more resources

@jpaulreed #Failoverconf

On thoroughness: We’re probably all doing this right now, by the way. @jpaulreed #Failoverconf

[This is actually a really interesting indication about our cognitive capacity and what workplaces are demanding of us during this time]

The first two are tactical and the last two are strategic. That is, the first affect you and are generally OKish for you to pull those levers. Shifting work in time and recruiting resources impacts the organization as a whole.

@jpaulreed #Failoverconf

Teams that do well with impact often have a strong Common Ground.

- Basic Compact
- Goal Alignment and Commitment
- Interpredictability
- Sustain and Repair

@jpaulreed #Failoverconf

It’s important to know when folks LEAVE the basic compact, especially. “I am not available, I have to take care of my family” needs to be said.

@jpaulreed #Failoverconf

<video of a plane landing on the highway>

Important takeaways:
- When did you go “this is not normal!”
- Some cars turned on hazards, why? Would autonomous ones do that?
- No blunt-end, beaurocratic intervention, all people at the sharp end.

@jpaulreed #Failoverconf

So remember, It’s not governments or policies or mandates that are going to get us through this. It’s people. It’s us. We’re only going to get through this together.

@jpaulreed #Failoverconf

[Community Building = Resilience Engineering. At me.]