I think this is great advice. "The problem with our current system is that the people running experiments are the only ones who can study them." writes . "y providing the community with more transparency about the experiments, the teams in charge of them can establish best practices for making decisions and reveal effects from experiments beyond what the team is studying." This breaks down into two types of openness:
- Open-source methodology: What is the intent of ranking changes?
- Open-source experimentation: What are the consequences of ranking changes?
Viewing the algorithm on its own tells us almost nothing about the software. But knowing how it's being tested and what testers are looking for tells a lot about what it's being designed to do. And, really, is what's important. Image: Towards Data Science.
Today: 5 Total: 116 [Direct link] [Share]
