Skip to content

Instantly share code, notes, and snippets.

@heathd
Last active October 6, 2017 07:52
Show Gist options
  • Save heathd/e87e6d65153f1c591b3197ae5715e797 to your computer and use it in GitHub Desktop.
Save heathd/e87e6d65153f1c591b3197ae5715e797 to your computer and use it in GitHub Desktop.
Mapcamp 2017

julia pierce

mapping->duplication (post merger) "we'd done this to ourselves"

separation of users into two classes affected everybody

liam maxwell

government had designed itself the wrong way

tech works horizontally open is a tool of disruption

  • open standards
  • open data
  • open markets
  • coding in the open

change needs spend controls (forcing functions)

square of despair:

  • procurement/security/capability/legacy

"agile"

system error report lead to cargo-cult agile

great work (Jamie's agile scaling sketch) was lost to poor message control

agile is a discipline

hosting

  • government thinks it's really big. £1.6bn/yr

Crown hosting

business case was to save £100m in 7 years but we saved £1.5bn to date

What's next?

we used open to unlock change

mapping gives us confidence to change and deploy the most powerful force

true competition in an open market delivers

open source

government doesn't have competition

co-operation and collaboration

open source tooling

minimum viable agency

  • paas
  • registers
  • open source components

Influences

  • Sun Tzu's five factors
  • John Boyd OODA loop

The importance of Why?

  • why of purpose (moral imperative)
  • why of movement (rationale for choices)

it's through movement that we understand the landscape

Vikings

  • navigate using stories (not maps)

Chess

  • sequences of moves
  • if you can see the board you win

Thermistocles

http://today.tamu.edu/2015/10/15/ceo-effect-on-firm-performance-mostly-due-to-chance/

Mapping

  • commoditisation enables higher order systems
  • the adjacent unexplored
  • inertia to change
    • compute; we don't like change
  • eg. blockbuster, netflix
    • who was first?
      • with a website: blockbuster
      • ordering online: blockbuster
    • what held them back? Late fees from stores

online image manipulation

  • we could use the maps to know where to invest
  • also know what to ignore

types

  • activities; practices; data

co-evolition

  • we used to build applications based on the concept of servers as a product; high MTTR;
  • architectural practices co-evolve with technological evolution

eg. best architectural practice for legacy

serverless

punctuated equilibrium

  • exponential growth

Doctrine

they're context specific

flanking is context specific

e.g.

Doctrines:

  • focus on user needs
  • reduce duplication (share maps)
    • in government 118 workflow systens
    • in global finance 1000+ risk management systems
    • in global defence company 2000+ accounting systems
  • Bias (outliers in evolutionary stage)
  • use appropriate methods (depending on evolutionary stage)
    • most companies yoyo because they don't have maps. Let's use agile everywhere! Let's use six sigma everywhere! Let's outsource everything!
  • think small
    • break things down (FIST Fast independent small tiny)
  • use small teams
  • different attitudes: pioneers, settlers, town planners
  • design for constant evolution
    • boiling frogs GCHQ

Leadership

Context specific forms of gameplay.

  • Flanking vs firing

eg.

  • open source - drives to commodity
  • FUD - slows down commoditisation
  • exploitation of constraints

Gameplay:

  • Innovate leverage commoditise (ILC)

    • turn something close to commodity into a utility service through API
    • componentisation effect (people build on top of me)
    • mine the metadata (I can see inside of the people building on top of me)
    • identify new patterns which I commoditise
    • I get everyone else in the world to innovate for me
    • if I want to be really brutal I use a press release process
      • you can't build anything new in the company until somebody has written a press release => the ever growing platform why does it matter?
      • innovation rate based on size of ecosystem
  • zimki

better for less; mapping at nation state competition levels

Back to the Phoenix project

(the case study used for mapping, nothing to do with the devops book)

Market development (costs money)

Doctrine sheet. I go through and look at competitors whether they are good or bad at com [photo]

CIO seems to understand but they're trying to get rid of her.

Paul Shetler

  • iron law of beureaucracy
  • deskilled workforce

mike bevan's square of despair

  • inappropriate procurement
  • inappropriate IT
  • inappropriate funding
    • business case
  • inappropriate governance

inertia is all that exists:

habits -> culture -> policy -> law

put a senior minister in charge

change management or change the management

work with human nature:

  • use the natural competetiveness of aspiring career managers
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment