Commit 4b04cbf5 authored by Jamie Tanna's avatar Jamie Tanna

WIP

parent 07dbfe1f
Pipeline #107142250 passed with stages
in 5 minutes and 32 seconds
---
title: "DevOps Notts November"
description: ""
tags: []
license_code: Apache-2.0
license_prose: CC-BY-NC-SA-4.0
date: 2019-11-26
slug: "devops-notts-november"
draft: true
---
# The Oauth of Non-Allegiance
Why "one true way"?
steal ideas, pick-and-choose, **??**
Don't change things when you don't need to!
Get right people in the room (for CAB) and ask them if there are objections to deploy
take ideas where good i.e. full-team planning as a team (quarter's work PI planning)
dependencies, risks
CD book from Jez + David
test automation - optimistic cascaed etc (unit tests on check-in, then integration, etc)
anything goes wrong? y'all jump on it
theory of constraints - need slack to allow folks to **??**
allow folks to get things out at a good rate, rather than bottlenecks
find your bottleneck, and try to eliminate it
queueing theory - how long is something stuck in a queue, with no one using it?
measure throughput, cycle time, lead time (how long intent in team -> being worked on -> done)
make visible where possible
lean startup
jurassic park - doing things because can - don't waste money, don't be bothered about "on time, budget" if not the right solution
XP - used to be extreme, now standard practice
pairing!
logically it should be slower if you have two people on two different things - but it doesn't work out that way!
great way to stop interuptions
pairing for not just programming!
cynefin - "habitat" framework
simple - helpdesk, very structured set of what to do, flowchart
complex - need people, not immediately obvious what it is
chaotic - prod is dead, fix first, then investigate?
understand where you are?
transformation management - `Our Iceberg is Melting` / meerkats version
have a vision of where you're going
don't know where from / to - gonna have issues!
pomodoro - multitasking sucks
monte carlo estimation - best case / most likely / worst case
go through mathematically and simulate expected? (lots of time, may not be more accurate?)
but shows not just planning poker
don't spend all the time on planning as it detracts from delivery
checklist manifesto - pilots don't take off without a checklist, stupid errors
proven with surgeons that it saves lives
automate checklist if possible!
sunk cost fallacy - you've already spent it, why not think about how much you can save/trim with new project?
time/cost is already gone
tech debt is ok, it's the technical compounding interest that will kill you
five whys - be a child
get to the point where someone blames a person, but then a couple further will get to the real root cause
don't be afraid to ask the stupid question
choice paralysis
nudge theory - "90% of people on ${your_street} pay on time"
make things hard - people become experts on getting around it
intent-based leadership - people on the ground know better than management! don't tell them what to do, show the vision and let them take it from there
people!
pick-and-mix what works for you, steal bits from everyone
#
don't follow the "elite leaders" because you'll never get to be them
find bottleneck, remove it
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment