Illustrating Pareto Principle in Automation in Testing
What I was showing Jon is a simple mathematical argument that it is better to "automate" in testing by focusing on "biggest bang for the buck" rather than "low hanging fruit" (low cost first) or "Biggest win" (high value first). You should consider cost vs. value on a case-by-case basis, which will give you the so-called Pareto solution (20/80). However, on a different level, I was showing him the difference between doing a spreadsheet "manually" vs. using Co-pilot. Note that, though Co-pilot did okay at the task, we stumbled into a big bug-- a bug that I suspect Microsoft doesn't know about because this is the sort of thing that is difficult to automate... The bug was that when we set Co-pilot to performing a task, then navigated away to another Excel window and then navigated back, we found it not only stopped performing that task, but lost all of the context and transcript for everything that cam before. Damn it, Microsoft! Then, when we tried to ask Co-pilot about saving the history, we found that there is no automatic or in-built way to save the history automatically. WHY DO AI VENDORS THINK WE DON"T EVER WANT TO LOOK BACK AT HOW THE WORK WAS DONE?

On Magic Boxes and Magic - Conversations on Taking Testing Seriously

How to Report AI-Assisted Work

Humans In The Way #1

Live Testopsy with Jon and James Bach

James Christie Conversation regarding Testing and Audits with James Bach and Michael Bolton

What is Normal? - Conversations on Taking Testing Seriously

Conversation #1: Checking, Testing, and Mutation

Experienced Excel Users Are Replacing IF With These Functions

Why birth rates are falling everywhere all at once | FT

Why Aliens Would NEVER Invade Africa

How SpaceX Humiliated Wall Street

Technical Work is All About Learning - Conversations on Taking Testing Seriously

Domain-centric? Why Hexagonal- and Onion-Architecture are answers to the wrong question @ Spring I/O

LIVE: Conan O’Brien speaks at Harvard graduation ceremony (full)

Co-Creator of Haskell: Functional Programming, Thinking in Types, Useless Languages | Simon Jones

Why The Russian Accent Terrifies Everyone

Inside Anthropic, the $965 Billion AI Juggernaut | The Circuit

Is the AI Boom About to COLLAPSE?

Stop Prompting Claude. Use Karpathy's Method Instead.

