Video: The Future of Software Engineering - Mary Poppendieck

April 7th, 2019

[ ] Book: Antifragile - Nicolas Taleb

[ ] Paper: Resilience Engineering: Learning to Embrace Failure - Acm Queue

[ ] Book: The DevOps Handbook - Gene Kim, Jez Humble, Patrick Debois, John Willis

[ ] Book: Continuous Delivery - Jez Humble, David Farley

[ ] Paper: Online Experimentation at Microsoft - Kohavi, Crook, Longbotham

[ ] Paper: Standis Group Study Reported at XP2002 by Jim Johnson, Chairman

[ ] Book: Sprint - Jake Knapp, John Zeratsky, Brad Kowitz

[ ] Video: The Design Sprint - Google I/O 2014

[notes] Video: The Future of Software Engineering - Mary Poppendieck

Video: The Future of Software Engineering - Mary Poppendieck

https://www.youtube.com/watch?v=6K4ljFZWgW8

 

 

 

 

1) scale out

more servers

CAP theorem

less communication

autonomous teams

 

 

 

 

2) infrastructure as code

conway's law: org structure == architecture

autonomous teams: independent deployment

ops: self-service

there is a cloud in your future

cheaper

more stable

more secure

more expandable

infrastructure as code

containers

serverless = event-driven

software-defined-networks

 

 

 

 

3) The New Technology Stack

dependency problem

no description for image available

 

 

api > db

no description for image available

 

 

big data

no description for image available

 

 

resilience engineering

no description for image available

Book: Antifragile - Nicolas Taleb

Resilience Engineering: Learning to Embrace Failure - Acm Queue, September 2012

 

 

 

no description for image available

 

 

 

 

===

 

 

Devops

no description for image available

Book: The Devops Handbook

Flow

Feedback

Experimentation & Learning

 

1) Flow

no description for image available

Book: Continuous Delivery - Jez Humble, David Farley

 

 

deploy all the time

release is flipping the switch

 

 

 

 

 

 

2) Feedback

no description for image available

 

 

no description for image available

Paper: Online Experimentation at Microsoft - Kohavi, Crook, Longbotham

Paper: Knowledge Discovery & Data Mining - exp-platform.com/expMicrosoft.aspx

 

 

no description for image available

Paper: Standis Group Study Reported at XP2002 by Jim Johnson, Chairman

 

Model: Problem Solving Teams

no description for image available

 

 

 

 

 

 

3) Experimentation & Learning

 

 

no description for image available

 

 

  • start with signals, not requirements

  • have a problem statement, not features

  • plan with hypotheses, not estimates

if you have good problem solving teams, what good are estimates

within constraints

here is my constraints

here is my problem

experiments around a hypothesis, not backlog of stories

analysis and conclusion, not guesses

 

 

Model: Design Sprint

The hard part is:

What problems are we gonna solve?

no description for image available

Book: Sprint - Jake Knapp

prototype and test any idea in 5 days

before actually coding

 

 

 

Video: Design Sprint Google IO 2014

 

 

Model: Idea Generation

idea generation (Tuesday)

no description for image available

 

 

Conformity Bias

no description for image available

 

 

Wednesday: they vote

-> don't like

-> needs majority

-> Attract people to ideas

if you can attract a few ppl to an idea

 

 

 

 

===

 

 

SUMMARY

no description for image available

 

 

 

 

===

 

 

Q&A

Site: www.poppendieck.com

 

 

 

Trend away from monolith and towards distributed systems.

Try and solve problems they don't have.

Over-engineer solutions.

-> Speed of learning

-> Need to migrate

-> Find your own solution

hadoop took 8 years

amazon took 5 years

if you wanna do it in 6 months, you're crazy

 

 

 

 

What do you think about the role of regulation systems like laws and government in the cloud federated world.

-> there will be regulations

-> cloud ppl better figure it out

-> cloud can be in many different places, and the location where will change rules

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Video: The Future of Software Engineering - Mary Poppendieck

 

 

GOTO 2016 • The Future of Software Engineering • Mary Poppendieck

https://www.youtube.com/watch?v=6K4ljFZWgW8

GOTO Conferences

58,251 views views

Published on Nov 26, 2016

 

 

no description for image available

 

 

 

 

Description

 

 

This presentation was recorded at GOTO Berlin 2016.#gotocon#gotoberhttp://gotober.comMary Poppendieck - Author of 'The Lean Mindset: Ask the Right Questions' ABSTRACT 2020 used to be far in the future. Today it’s four years away. We no longer need to guess what breakthroughs await us in that magic year, the future is hiding in plain sight: a reliable Cloud, industry-disrupting Platforms, massive data [...] Download slides and read the full abstract here:https://gotocon.com/berlin-2016/prese...https://twitter.com/gotoberhttps://www.facebook.com/GOTOConferencehttp://gotocon.com#ComputerScience#disruption#cloud#IoT#futurology

 

 

 

 

Top Comments

 

 

Steve Bar Yakov Gindi| at 42:15 "We need to move from Delivery Teams to Problem Solving Teams" Best MOTTO for the year!

2 years ago

 

 

Alex Ellis| Looks like an interesting talk. Anyone else notice that her arm keeps disappearing?

2 years ago

 

 

Hong| if we can see the slides more often, it will be great, or at least put slides on half of the screen, what is the point of showing the speaker all the time with a green background?

2 years ago

 

 

Aaron Hawley| GE sold off Predix. Maybe that's not a failure, and maybe Amazon has monopoly power in Cloud, but the sense was that it was a failure.

1 month ago

 

 

Ricardo S| amazing talk ... a lot to think about...

2 years ago

 

 

M Rizky| very inspirational talk - need a lot of things to learn I guess

6 months ago

 

 

Arseni Shylau| wasn't it Scale Up moves to Scale Out (in the first half of the speech) and then in the final summary slide Scale Out >> Scale Up?

2 years ago

 

 

Master Miscellaneous| Great talk. Really enjoyed listening.

2 years ago

 

 

Bryon Lape| Stimulus, Response...I swear some guy named Mills came up with that in the 70s.....

1 year ago

 

 

Dan Shepherd| With regard to the cloud based architectures. The main problems I can see is that sometimes the API's can change or even go out of business - I had that with using Parse.com in a number of mobile applications, all of a sudden the back end had to be replaced for a number of applications not through my doing. Luckily they open sourced their system, so I was able to just host it elsewhere and configure it again, but this is not really ideal. The other thing is cost, it may be cheaper to use cloud services but these are normally rental rather than one off purchases so for smaller businesses this can be more expensive in the long term - so there is still dependency - its dependency on the fat cat, i.e google or amazon etc. What happens if you have a free application that by fortune becomes very popular? All of a sudden your costs go up because you have to pay for the cloud services to keep up with the demand, but still your application doesn't generate any revenue - what you are left with is software that generates dept.

1 year ago (edited)