https://www.agilealliance.org/resources/videos/xp-as-an-incentive-system-kent-beck-xp-2018/

Incentive System

• encourages some behaviours

• discourages other behaviors

Actual Talk: puzzle & key

puzzle: why is it that we have a thing that works well & is enjoyable ; people wont do it

how people look at you refactoring depends on gender

on male dominated teams

• men who refactor → "craftsmen"

• women who refactor → "fussy"

anecdote: warm-ups playing games, practicing forming team

what does refactoring look like, if you're really good at social interaction?

what does code review look like,

to what degree are the development practices, as we practice them, attempting to make up for a deficit in, say social skills

how good could software development actually be, if we assumed that people would deliberately practice interactions

→ limbo- open space session

what if you wanted a 100.000 programmers working on the same system

could you take the traditional code review system and scale it up to 100.000 ppl

:( code review

• costs

• cost of waiting

---

extra talk: 3X framework

success = the ability to regret mistakes that you made earlier

no estimates ; it depends on what your risk is

• explore: nobody cares

• expand: you can't scale

• extract: you can't be sustainably profitable

---

Q: xp practices on the 3X

what you care about in the extract phase is engineering efficiency

other problems that are not discussed: architecture

architecture is a big cost driver - how to discover opportunities for big architectural changes & apply them - xp does not discuss that

waterfall was attempting to address extraction, but applied to exploration problems

→ year long projects, nobody using it

XP contains a lot of sustainability → may or may not apply to exploration

Q: elitism

"I have all this power" largest lever that ever existed, reach billions of ppl

→ responsibility, etc

"therefore I am better than others" bullshit, lack of self esteem

→ take care of that in private, where it can't hurt others

Q: facebook model

if you can make experimentation cheap enough, then you don't have to think about whether an idea is a good idea or not, you can just try everything

the need to shift focus when you got into expand, and bottlenecks start appearing

Q: Andella (African group that wants to 100.000 world class developers)

is their target achievable?

→ exceed their goal, numbers

→ going to redefine what it means to be a world-class developer

-→ social interaction & teamwork

-→ what if the expectation was "oh programmers, those people who get along with other people"

how does it apply to 3X ; where would they

→ different people specialise in different parts of the curve

Q: reduce cost of experimentation

red flags of increase the cost of experimentation

XP did not do "role of visual/interaction designers" well

• extract: important to do careful design

• • small-problem * lots-of-users = big-problem

• exploration: good enough so that you can trust the results of an experiment

• it does not follow the corporate standards for

• how is this going to scale?

• • → it's gonna scale because I'll delete it on Friday

• basing conversations on fears

• • → I want the answer now

Q: Security in 3X

if your product is small, there is less risk

the amount of effort, changes across the 3X

Q: Discrimination against more junior

I'm fine with discrimination against incompetence

tackling discrimination

step 1: know that it exists

my way is easy, because it's the default

acknowledge it, if I'm the senior - my default position is: I know and you don't

→ but let's challenge that

pattern for junior-senior pairing: give homework

we're gonna use this API, I don't know it, go study it and then teach me

they have something to bring, I have something to bring

Q: explore: lean startup? vs 3X

:-( customer development = "people in a room think really hard about what customers like"

lean startup matches in explore phase

XP is really good at

idea, how will users behave? → oh this is how users behave

that is the only loop that matters in explore

:-) nobody would push that button → how fast can we put the button there?

:-( let's try to make sure people try to push the button

XP team should be able to answer "how will users behave"

→ 4 months from now, answer such question in half the time

take-off: change from "latency focus" to "scaling & bottleneck focus"