Compare table: should we normalize values?

[arouault]

This is more of a question but should we be normalizing the comparison table?

Right now it reads like "Ames spends 5900% more than Acton" -- this doesn't mean much to me as a resident.

Is this a lot? Is this normal? Is this just representative of population/size of the city?

[ericyoungberg]

This direct population-based approach won't make sense. The population of the town or city would be irrelevant for Commercial and Industrial sectors. This approach will only make sense for Residential.

Perhaps we can use the following to perform the text analysis for each sector:

• Commercial: by averaging the consumption per worker column for each muni
• Industrial: by multiplying the average employee by the number of establishments for each muni

This will include some additional munging so I would rather wait to see where we are at with the other features first?

[ericyoungberg]

I have to remove the code for now to fix #28, but when I return to this issue in the future, I can find the population-based normalization code in commit b297402f5fbc491a8ddde631792d288577ec466d in the components/spider-chart-layout.js file.