You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If we want to use age as a dimension w/o breaking our pre-aggregation approach, we probably need to apply binning to collapse the integer (fixed precision 1/100th year) into a smaller number of discrete values combined w/ other vocab term concepts to identify a "core_fact".
However, should the bins be uniform or non-uniform intervals? Different bins depending on human or not...?
The text was updated successfully, but these errors were encountered:
only display human data: nobody wants horse age shown on the same axis as human age (advanced future versions can get a species selector)
nonuniform bins based on most frequent approximate divisions of (human) developmental research categories: blahblahnatal, adolescent, 18-29, some Nielsen demographic or two, 65 & up, etc. -- find a big, well-publicized list and copy it
also an alternate display option using uniform bins based on decades (0-9, 10-19, etc.)
...and we're in a void right now, because i don't know who (if anyone) will be submitting (public!) age data to us this quarter, and so cannot yet say how it will look, let alone how it should look
is there some well-defined option (ideally with a precdent) for "do not include a selector for this dimension in this release because we don't yet have any (or enough) incoming data for the field?"
If we want to use age as a dimension w/o breaking our pre-aggregation approach, we probably need to apply binning to collapse the integer (fixed precision 1/100th year) into a smaller number of discrete values combined w/ other vocab term concepts to identify a "core_fact".
However, should the bins be uniform or non-uniform intervals? Different bins depending on human or not...?
The text was updated successfully, but these errors were encountered: