🧶THREAD - Un programme de 15 lignes de code Python arrive à rivaliser avec les meilleures intelligences artificielles !
Cette drôle de découverte vient d'être publiée par une équipe de chercheurs canadiens, et risque de bouleverser le monde du Machine Learning.
Explications ⤵️
Code-review-as-inspection is ineffective. Don't just inspect, pair or mob with the person who wrote the code and fix the problems. Right then. No delay.
Ofc, if you had done that when the code was fist written, you wouldn't need the review. 😄 Quality cannot be inspected in.
You WILL make architectural errors early on in a project. Every time you realize such a mistake but don’t correct it, you move one step closer to hell.
I'm no Scrum guy, but ISTM you pick a Sprint Goal and then do whatever works to move you towards that goal. What you do doesn't matter much. Given that, I don't really understand all the obsession surrounding estimating stories. Why bother? Pick something valuable and work on it.
@everton_dev@klmitchell212@nicklockwood Apply SOLID principles instead. Clean Swift is based on Clean architecture with a rigid approach. You don’t need 1 View controller, 1 presenter and 1 interactor all the time. I found very useful to read Clean Architecture book by Uncle Bob to avoid these boilerplates.
@everton_dev Gather together those things that change for the same reason, and separate those things that change for different reasons. Make sure your app is testable (dependency inversion) and your tests are fast enough (< 5sec). Don’t follow one particular pattern, be pragmatic.
@k_ebrahem_ @AmiraultTony Let’s do a part 2 and go more in dept. For those who are interested about clean architecture, we will release the conversation in video soon.
Back to learning about ASICs. I will post learnings and progress here as I go. This is all just for fun to help improve my first principles understanding of hardware to help with @MightyApp down the road.
The primary strategy of architecture is the drawing of hard boundaries between high level policy and low level detail such that the high level policy is entirely ignorant of the low level detail.
It is clear to me that test last is a bad idea compare to test first.
Starting with a failing test prove the absence of the requirements.
Adding your tests after your production code can't prove the presence of the requirements. Unless you change your code to make test fails.