What is the problem you are seeing? Please describe.
Apps can be evil but still be rewarded by app mining.
How is this problem misaligned with goals of app mining?
Can't Be Evil is a core value.
What is the explicit recommendation you’re looking to propose?
The final score should be calculated by adding the review scores except the CBE score, then multiply the sum by CBE. As in:
(UX + Marketing + ....) * CBE
Describe your long term considerations in proposing this change. Please include the ways you can predict this recommendation could go wrong and possible ways mitigate.
CBE will have major impact on the final score. The CBE reviewer(s) will have overriding impact on the ranking. Scores should be normalized to be positive before multiplication.
What is the problem you are seeing? Please describe.
Apps can be evil but still be rewarded by app mining.
How is this problem misaligned with goals of app mining?
Can't Be Evil is a core value.
What is the explicit recommendation you’re looking to propose?
The final score should be calculated by adding the review scores except the CBE score, then multiply the sum by CBE. As in:
Describe your long term considerations in proposing this change. Please include the ways you can predict this recommendation could go wrong and possible ways mitigate.
CBE will have major impact on the final score. The CBE reviewer(s) will have overriding impact on the ranking. Scores should be normalized to be positive before multiplication.
Additional context
191 Can't Be Evil sandbox
209 Weight of NIL Dimensions