Currently the utility measure only compares count(distinct uid) between anonymous and raw data.
I would like to add a second measure, which compares count(*) between anonymous and raw data.
This change will require everything from the .json specification file to the final output graphs be modified.
In so doing, keep in mind that we will have additional measures over time, and try to make the changes in such a way that new measures are as easy as possible to introduce.
As a first step, please describe how you think the changes should be designed. Once we have a design we agree on, then we can start coding.
Currently the utility measure only compares
count(distinct uid)between anonymous and raw data.I would like to add a second measure, which compares
count(*)between anonymous and raw data.This change will require everything from the .json specification file to the final output graphs be modified.
In so doing, keep in mind that we will have additional measures over time, and try to make the changes in such a way that new measures are as easy as possible to introduce.
As a first step, please describe how you think the changes should be designed. Once we have a design we agree on, then we can start coding.