By the time we extracted Scientist, the code we were refactoring had pretty good test coverage. But even the best test suite is an imperfect model of production. The first section of the README briefly mentions this:
"Let's pretend you're changing the way you handle permissions in a large web app. Tests can help guide your refactoring, but you really want to compare the current and refactored behaviors under load."
Added: That's a little too glib, sorry. Let me try again: From my perspective, if there's something that needs to be done at GitHub there are a few possibilities:
1. It needs to be done and it's getting done,
2. It needs to be done and it's not getting done, or
3. It's bullshit.
Cases of #3 become obvious pretty quickly. The best evidence: Searching for ways to make someone do it because nobody stepped up.
Cases of #2 can happen for a bunch of different reasons, but malice, apathy, laziness, or incompetence are the least likely ones. The most likely: Not enough hours in the day or not enough people with the knowledge necessary to be worried. No matter the reason for #2, someone at GitHub who is worried will generally try to get others to share their priorities, by persuasion, by hiring, or by prototyping.
Or occasionally by just jumping up and down and wailing.
We use PRP in any situation where someone explicitly takes responsibility for an outcome, no matter whether the outcome is some software or a clean office or an accurate tax return.
"Let's pretend you're changing the way you handle permissions in a large web app. Tests can help guide your refactoring, but you really want to compare the current and refactored behaviors under load."