Video: Is TDD dead? 5: Answering Questions

April 5th, 2019


Is TDD dead? 5: Answering Questions







What is an example of an open source project that does TDD well?




common code

open source only extracts commons, libraries

only personal experiences of private apps




takes a lot of effort to understand the code bases where these problems occur

team of 12 for a year

when you're not working on it




JUnit project




Are we scientific?

virtually impossible to true/false

it's still worth debating




You can't repeat because too many variables

scientific mindset is still helpful

no universality






What could change about the way we write software that would make TDD redundant for Kent and Martin?

What could change about the way TDD is performed, to make it useful for David?




blogpost: RIP TDD

problems that are solved


piecemeal progress

always having byte sized chunk next

overwhelmed -> spotlight





applicability in different contexts

low stress, rapid unhurriedness



no TDD when web UI




nasty language

unfamiliar language




started with TDD

free regression testing

regression testing is awesome

apply this to everything

kept trying to apply it everywhere

can't use TDD for the majority of things I do

when I see others having forced it, I see bad code

good feelings from TDD

regression suite

not design driven part




natural learning:

try it

use it

over-use it

dial back




"Oh if I had an X, then this problem would be easy to solve"



make the problem easy, then solve the easy problem

but making a problem easy, can be hard



looking for feedback - F5






How do you think a lack of design experience affects TDD?

How do you think TDD works with an inexperienced developer?





forces byte size pieces

forces separate interface from implementation, maybe not good design

produced self-testing codebase, can refactor


better design decisions




TDD is the gateway-drug for regression code and self-testing code

enlightenment to testing

no nuance discussions

fear of shying new ppl away




Dogmatic :-(

we need to keep repeating beginner stuff

new people are joining




mutation with everyone adding their own stuff

we need to reset to the beginning principles




reboot ok

dont wanna lose: expectations about software dev have changed

feel confident

point to my contribution

constructive collaberative




ruby: fun, hapy

now taken for granted




meaning get's lost over time

different set of problems




accumulation of cruft

leave good ideas the fuck alone

you're making good things worse by adding to it




build from the ground up every time

from first principles




"how did people get to this?"




In 10 years you'll not be able recognise what this turned into




Keep repeating the principles

Mention the added things that are useful




ruby is a reboot

functional is a reboot

nodejs is a reboot

rails is a reboot



success looks like being a villain




other avenues to same goals

is it just a matter of design where we're lacking




Fear of public saying "It's not working for me"




feeling comfortable in making changing





self-testing in code

TDD is useful in some context

disagree on how many?


your team

find out, use, over-use


This post was referenced in: