What the best performers will be doing in 5 years and why: a fantasy story

Thomas Depierre

Why try to predict the future ?

New Practices

Innovation

Wait 5 years

Best Performers Practices

2 to 10 years

Practice we implement

Edge

Lambda

No Container

Nix

DX

SPACE

...

Agile

Lean

DevOps

K8s

Containers

O11Y

SLO

Hopefully

The Plan

Agile

Lean

DevOps

Containers

15 years

How can we be faster than 15 years ?

What should we do today to meet our goals in 5 years ?

How to predict the future ?

Static Systems

Safety Science in Complex Domain

Safety II

Safety Differently

Resilience Engineering

Cognitive Systems Engineering

We build Complex Systems

Complex systems are systems whose behaviour is intrinsically difficult to model due to the dependencies, relationships, or interactions between their parts or between a given system and its environment.

fault tolerance isn’t composable

Peter Alvaro

Fun >>>>>>>> Business Critical >>>>>> Safety Critical

Reliability

  • Doing what is specified
  • Repeatedly
  • TDD, Type System, Proof, etc

Safety

  • Avoiding Loss Events
  • Financial, Assets, Human life
  • Systemic property
  • Not a technical problem only !

Humans create safety continuously through normal work

The Rasmussen Model

Drift into Failure

We build Dynamic Systems

Law of Stretched Systems

every system is stretched to operate at its capacity; as soon as there is some improvement, for example in the form of new technology, it will be exploited to achieve a new intensity and tempo of activity.

Larry Hirschhorn

The above-the-line/below-the-line framework

Mental Model

A few more models if we have time

  • ETTO
    • Efficiency-Thoroughness Trade-Off
    • if you wait until you have found the perfect solution, the system changed too much
    • better go with a partial one for now
  • Work-As-Imagined vs Work-As-Done
    • Incident is your system telling you where these two mental model diverge
    • Incidents as untyped pointers
  • Why do things go right ?
  • Incidents as unexpected investments/probes/experiments
    • you paid the price already, at least get some sense-making out of it
  • ...

Where to go for more ?